{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "13677315-a0c2-46bf-9fa1-7acece553a42",
   "metadata": {},
   "source": [
    "## 1) Plot all rmse results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "a73bb78f-fab5-4209-b166-3e07fe8ec303",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Models used: GPT-2 (base), LSTM, DeepSeek-coder-7b, BERT-small\n",
      "Saved to C:\\Users\\jacob\\OneDrive\\Documents\\GitHub\\external_data\\nlp4neuro\\results_april30\\experiment_1\\experiment1_rmse_sig_final.pdf\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAB+AAAANVCAYAAABI+ec5AAAAOnRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjEwLjEsIGh0dHBzOi8vbWF0cGxvdGxpYi5vcmcvc2/+5QAAAAlwSFlzAAAuIwAALiMBeKU/dgAA0i1JREFUeJzs3QeYXFX5OOCTkEAgSBLCBkFIaGII0oNKB0EUESV0BKQo8ENRA9JRihQhoISigIiAgig1IhYQISCCYpAqiBqBUAJZhCQkAULZ//Pd/zPr3cmWmd1pu/u+zzOQKXfumbJzzz3f+b4zoKWlpSUBAAAAAAAAAD0ysGebAwAAAAAAAABBAB4AAAAAAAAAKkAAHgAAAAAAAAAqQAAeAAAAAAAAACpAAB4AAAAAAAAAKkAAHgAAAAAAAAAqQAAeAAAAAAAAACpAAB4AAAAAAAAAKkAAHgAAAAAAAAAqQAAeAAAAAAAAACpAAB4AAAAAAAAAKkAAHgAAAAAAAAAqQAAeAAAAAAAAACpAAB4AAAAAAAAAKkAAHgAAAAAAAAAqQAAeAAAAAAAAACpAAB4AAAAAAAAAKkAAHgAAAAAAAAAqQAAeAAAAAAAAACpAAB4AAAAAAAAAKkAAHgAAAAAAAAAqQAAeAAAAAAAAACpAAB4AAAAAAAAAKkAAHgAAAAAAAAAqQAAeAAAAAAAAACpAAB4AAAAAAAAAKkAAHgAAAAAAAAAqQAAeAAAAAAAAACpAAB4AAAAAAAAAKkAAHgAAAAAAAAAqQAAeAAAAAAAAACpAAB4AAAAAAAAAKkAAHgAAAAAAAAAqQAAeAAAAAAAAACpAAB4AAAAAAAAAKkAAHgAAAAAAAAAqQAAeAAAAAAAAACpAAB4AAAAAAAAAKkAAHgAAAAAAAAAqQAAeAAAAAAAAACpAAB4AAAAAAAAAKkAAHgAAAAAAAAAqQAAeAAAAAAAAACpAAB4AAAAAAAAAKkAAHgAAAAAAAAAqQAAeAAAAAAAAACpAAB4AerHp06enIUOGpAEDBqTFFlssPfbYYx0+9pRTTskeV7hMnTq1pm3lf37+85+3fg4rr7xyWrBgQb2bBHTTu+++mx599NF0zTXXpAsvvDCdccYZadKkSekHP/hB+uUvf5n9Lr/11lv1biYA9Apbb711m3MWGkOcO+Y/lzi3pLqcvwPQW62yyiqtx6/4d2euvPLKNse7uN5XDKp3AwDovpdffjk98sgj6dlnn02zZ8/OBviXXnrpNHz48NTU1JTWW2+9tNJKK9W7mb3We++9l/79739nwZOZM2emuXPnpiWXXDItu+yyaa211kobbLBBGjx4cF3beMQRR7QGdvbZZ5+0zjrr1LU9lGbPPfdMZ599dnr44YfT888/n84666z07W9/u97Ngj4jTvDi2Fhw1113ZQP6lfS3v/0tff/730/XXXddmjdvXqePjWPFuuuumzbddNO0ww47ZG2J40lBnGAeeOCBqdqK34f49913373I44477rj0ne98p1v7iMHhbbbZpt37WlpauvWcAH3Jf//73/TXv/41Pffcc9k53MKFC9PQoUPTiBEj0qqrrprWWGONtOKKK9a7mQAAQA8IwAP0Mv/617/SZZddlm6++eYsONyVCMRvueWWWcDvM5/5TJsB/84ccMAB6aqrrur0MTErrRDwj8GijTfeOH3qU59K2267bbvZCh0N9FfSmDFj0jPPPNPt7V966aV04403pttvvz0LIkTQvSPxXu62225ZEDyC8bX2+9//Pv3qV7/K/h3Z76eeemrN20D3xN/Haaedlnbaaafs+jnnnJMOPvjgLBseaGxRseLYY4/Ngu+lBpTffvvt9OCDD2aXyJKfOHFiOu+881Kjimz+yOQfOLD8gmld9R2Ayk0uas8SSyyRXUaOHJne//73pw9+8INp7bXXTptttln6yEc+UvfJo/352BGTreI8LiZgdiU+u/i84rzq05/+dBaUB+qvs4mG3VGNSaJA/fph+bHSYcOGZf2wDTfcMO26665pk002qctvTbQjJvz15HXFeeEyyyyTPdeaa66ZNtpoo2w8KyaYtyfGZWOcuNpOPvlkFVloaALw0Eutctyv692EPuWZs3ZMje6FF15IxxxzTFa6OjKzS9Xc3JwFlOPyvve9L339619P3/jGN7KgeU9F4OH111/PLpHBcc8996Tvfve7WWfs/PPPz4LxvcnnPve5dOutt5b8/r7xxhvppz/9aRaoOPLII7NgxeKLL55q5Zvf/Gbrv/fYY4+adG6pnB133DF9+MMfTo8//nh68803s+/PJZdcUu9m9R+nDKt3C/qOU+ak/iL+VmOg4c4772x3oCUmgUWVlBigePXVV7Njd3vl5xs9EzyO6TEgHIGfcsyfPz/dcMMNVWsXXVOuuXIa/e+0I/GbE5eYRPr000+n+++/v/W+6P/vsssu6atf/Wpaf/3169rO/iTOkaLKyX/+85+yJgXfcsst2SXO32LgPAa9gb6lluMHVJcgYOX09vcyP1YaFQ/jvCrGSqPvFcuUlROIbxQxThp9kbhEsD4SgqKSY4xpXXzxxWnzzTevdxOhIQnAA/QCv/71r9MXvvCFbDC/PVGycLnllssuMeAWpemjtGFxIDk6f6effnq66KKLsg5TzF6shn/+859Zid3oNMdsxN7iT3/6U7vB98gUijKQUU0ggi8xeJZfszu2Offcc7PXHRMdBg2q/uE1Mt8feOCB1usxOYPeFyQ56qijsmoT4cc//nFW9rmrtZGA+vna1762SPA9skqjEkpMOovjcXHmeyxjEoMu119/ffa73V5Q75Of/GQ2iFGKqJgRVVoK9ttvv6yPUIpYmqYzMXGgcBz8yU9+UnYA/qabbmpTjj//fED9xaBp9DfiElWcJk+enD7wgQ/Uu1l9WkzujUkPcTwoFudio0ePzo4d8dsZE6dnzZrVpyaEAB0bNWpUVukC6B+iAs4WW2yRrrjiiuwcri+IhJKtttoqq/L25S9/ud7NgYYjAA/Q4CLDOjIm3n333Ta3RynJL33pS9ngeHvrfsdagpFt8dvf/jYLCufLCRXWGixHewP8MRA0Z86c9OSTT6YpU6Zk6+HmRQA+1qD/4he/mF2PGZ+vvfZaSfv7xCc+0eb61VdfnZZffvkutyu1xH5XYl/xeiOgEiWVhgwZ0npfDKD97ne/SyeeeGIWWCmIDJUIoEYwvtpi/fCCj33sY7KYeqlYGiICd/F3Ed+rGAiPC9B4HnnkkXT55Ze3ue2kk07qdPmPmMAVJQfjEtVnYhmZqBATJQnzVlhhhexSijge5q222mppu+22S5UQZQ7/8Ic/tAbTI0OjeFJBqeXnx48fn73e6CcA1RF9zuKJNdGfiH5FIUMpMuCnTZuWVW7Ki2oVUd40Jgcpf1wd06dPT7vvvnub4Hv8pkZG+7777pvWWmutdjPf//KXv2TnFXF+1dEEbKA+4je31EmTxWLJsfxyefvss09NJu8DteuHhRi/jUpETz31VDYme99997W5L5Iwxo4dmy3jWc4YZfF5YKnKXX6ovdcV7Y7+ZYyBRh8yEpAKYsJ1VFdaffXVs4nlhaV0Sv2tjMnlMcm8YN11183Gj0sR58LQyBzlARpYZMpF8DoffI/SkRdccEF2stbZ2qxRyiwG5OMSZYFizcEocV3KekXt6WqA/1vf+laWLXfIIYe0Kbcb6+TGwFNkeMQaQd0VGYa1yAyO8kmRtb/zzjt3eDIcndcoQRyTBOK1RWZLQXw28R5EGf5qiTWEI1s/fyJP7xQTO+JvOapShJgJHWvDx3IRQONNiMtnc8fgQmfB9/bEGoCFv/dGFBPPCgH4yMaMCXylZtcXytbnnyv6BkD1RN+6lOB5YdmkmOQXE2cLXnnllWx98RgcjuwlKisqVEX1rIJYLioGo2OAuiMxYB3LYsUllia67rrr0nnnnVejFgNdGTFiRLcmPkawKh98D4VKaEDf7YfF0pHRz4plIwuVwuKcMqoh3n333WWNHVVq0nVPXtdee+2VVVaNAHn0cwoVeuI1xYTzGCeNsepy2htl+ivxOwuNqOPIDQB1FdkO0UHLZ0xENvm9996bZaN3FnxvL2AcQdqYfXnYYYdVqcX/f7C9OLAQpfAjMN8bRGZjZDhGSc5SZqJHh/LnP/959rkUxOeVzwCshshILFhiiSWySQD0XvngVsySvuaaa+raHqB9xTP4Dz300NTXRLZBvqJKOcfv/ASF6HfsvffeVWkjUL6oEBUTRB999NGs8k5xcD76kjNnzqxb+/qiyBKLLPa8n/3sZ50G34vFb2lM1IwKBjEJG+i9IiEib4MNNsj6XUDfF0t05sfxwh//+Mde2/cqLKcYl7y///3vbbL9AQF4gIYV5dvz2eoxABODOFF6vrsiWBudvigXFBny1RAZ+x/60Ifa3JZfq7aRRaZJORMbCmUkY03gvNtuuy1VS2TRxOdXELNLZUv3blGmOT+Jo7tlxYDqKp6ZH2UD+6L999+/9d+R0V78ujuSn3y24447puWWW64q7QO6LyaYfu9738suebH2+NFHH123dvVFUd7/nXfeab0egbZYNgrof2KCYkxUzJP9Dv1LTE4eNmxY6/XIHI/103uzE044YZGx5UI1NeD/U4IeoAFF1njxOrPRsYlZ0pWw6667pmrOhIxSlpFtX5BfJ70v2mKLLdpcnzFjRtX29etf/zrLks5PGqiU119/PZuFG2v2zp8/PzU1NWVLD2y55ZZlrxlVLAI4MRv26aefbl0PeNlll00f+MAH0iabbJKVmOqpmLDy0EMPZfuK9yi+izFBItZVjtcRk1eiakF3xVqqMZs3ZinHQHU8V7xH8Xc5bty4bj9vtPOzn/1s64zo2Ee8ljFjxnT7OYHKi9/IvPzyMH3J5z//+SwQF4GjGDCOSUHHHXdcp9v8+c9/brMOYKll64H6iCz46PPdfPPNbbKzY9mI4om0XYl+b2Rov/zyy1nWdwwuRxn1WD4q/l8J8Xsb+/j3v/+dZs2alS13FX2wKOke+4lJxpUU+4l12F944YXsevRXN9xww3bXbG/0SVvxfsVvdKwtH+eYSy+9dBo1alT6yEc+kr1/lRBBhKiuEMsbxP7iPCImYcUE0zhPin32dYXv6PTp07OlHaLPEK87vjtxnhDfnejzd8fDDz+cnnjiiey9jcnY8fmtvPLKafPNN8+qW/RUVHCLMsj/+c9/srbH+VMsmROfXTUmetfiO5lX+O2Ic7h4/+IcK/o6tRQBqfxvQpxbV6IN1Tx/Byo/CTJ+W+M3qSDGlXqzqM4TCSX5rPf4ve2t4nw2qqLG8SJ+X+Mzi2NiHMujglEcz0upltqV6CvE2GX0M6MSVfSX4nd79OjRXfa34vsT/YL47kTbYqnWj3/84z06Xkc/IMbwo68Rx+Z47fHcMWYb/dfoA1fidfdX3jmABnTxxRenBQsWtF5faqmlFikX2cjipC8vTuT7suLgcSHAXA359ebDNtts0+PnjI7biSeemJUajgHNYtHpilLLJ510UskB7Ajc3HHHHVm2fvw/X82hWAxGRUZQrB8VEwrKGZyK4NCPfvSj9P3vfz8b+OtMzMyNtaxiAsqXv/zlkgesfvWrX6Vzzz03O6nIZzLlRUd54sSJ2fN2ZxA4PsdCAD461THRIp4LaKwBhvwgSQwef/jDH059TQyCf+pTn2o93sSxoasAfD77feTIkVkGPNDYom/zy1/+snXpiOh/XHrppYtkx7dn4cKF2fnKD3/4w2ywrj3Rn4t+VwT1Y6Jhd8R6yaeddlqaMmVKtjxXe+I8aZdddknf/va3Sw7exWBloW8awbjCusyRtX788cdnv+/tWW+99dIZZ5xR0m9cPSdtFSZPXXjhhenBBx9sXZ+1WASFo/8dk6bKrQIW4pj4ne98J1uSq6MyutH/jmNKfI7VLrcdfemvfvWrrd/pxRZbLF1yySXpS1/6UtX2GROMzzzzzKzv3tk5YExGifOcWA4uvkelfH/OPvvsdMUVV6QXX3yx3cfEeVnhve1OfyQG/ePvJv6O2/v7inOa+G7E6+tpVZtqfSejrPuBBx7Yej3er8gsj9cWv3FxPSaA58UkoVoH4IvLz3/mM5/p0XtajfN3oPqKx7oqMYmq3vLVHHvj+G/8hk6ePDlddtllWWC8M/F5RQJRLN30f//3fyV9zltttVXWvwy//e1vs2NqLC3b3jY777xzuuCCCxZ5T+OYGe2LbdsbW43jdfR/Tj311KxfXIr4nK677rrsnD8mcs2bN6/Dx0Ywfq+99srGBNZYY42Snp//UYIeoAHFYFheHNzzpYoaXfFs646Cln1FITsmH3yolnw5/8IszJ6I6gQxCBSdufZO3kMMyMTgWjyuMEDZleicxTpXUcmhs+B7oTN5//33pwkTJqTddtstm71fisiyilmiMbjQVfC9MFgc+4l1qoo/s46yMyIwHoPG99xzT6ff48j+OvLII7OBxa467e2J15H3u9/9ruznAKqreAmYs846q8OAUG+Xz2CPjMa//vWvHT42jh2/+MUv2vz+V2uZG6CyE2Z32mmnNrdFoLsrkRke2TAx8bCj4Hs+SyeCjtGXKrV/V3D66adn2fg//vGPO/2tjUnLEdiLNhVXECvHOeeck2UQdRR8D5EVFYGzGHTtKIBYULxm+wMPPNAaGK6myISNTKVYTiTe/87aGb/vEbyMKgLlZuHFex3nIeedd16na9hG/zuWUYuKURHsrZaoFveVr3yl9T2OAeg4p65W8D2yxWKybPT9o3pEVxOwI6MsJriUUr0sstFjgDsme3QUfA+RzR1/s+uvv34WjC1HBKXj3K6zvkwc3+McMR7Xk4p2tfpOFsS5Z2RkRvC5OPheD1GZLV9tpKfl56t1/g5UV0zEi9/DvL4QzCz+Te9utZd6iHG8OIZGYLmUcbyY3HXnnXdmk+nKHeeOfkpUi20v+F54H+NYEdVg8pXlYp/RX49xz47GVuNYEJPOPvnJT5bU346x1KgUGv2mmBTQWfA9xHNGvy8m+xVPKKNrAvAADSYOfFGKJq+3ZZIVDxRUMyDdCGK2YN6aa65Zlf1EhzD/3sbAQk9EqdDImsgPmsVMy3jeGJQtzjqITmAMTHY2EJQfECoWJfGiZNNHP/rRbDCgvVn/N910UzYw1dUAZXRO43F/+tOfFunsr7jiitkgT+wn9lc8AFqKODGKrPzCTNX880fWVLxH66yzziIlNeM9ihmx+Q5zKQqlJDv6TgH1V3wsjt/k+C2I4HMMxPclESzLV3eJDKuORGAlTuLbW0MeaGyROZ4XwarOJk5GVaCYnFgc1IpJNxEsj0HDCIQXl6mM7aIP2V7/sL0B6ghOReZ8BG+Lzyki4Bm/vTERNS8eG8HWCAiXK9ZmjqzbwiByZBLF64n+ZPRfi0UgtbPMp/YmbT333HPpoosuStUUkyM23XTTbKJAXmSCxyB/fD7RNy7OiI1JB9F/LTXgGZ9NvNfFWf7LLLNM9rpjP9Ffzou+/cknn5y+/vWvp0qK429MGotgY0F8ZnfddVfVzqEj2L7ddttlVSCKz1kiOy7OBeM8JLK5yy2/H5n0cX4WE4Hz4jMrlIEt/k7G30xkxX3xi18saR9xLhd/j8WBoML3JP6+4nwq//hoU5w7Nup3Mh/s3n777dtMDorzrMKSYZHFV2uRYRgBlHx7IgjTaOfvQHVdf/312bKGBfE7W9xX6I2Kl9xZfvnlU28Qv8txLP/HP/7R5vb4HY3qSFHFqdCv7elSOjHJM99PiTHKGBONS/FxKX7fIzkp+jfRx4gkpegbFETgPNoW353i/nYE92OCbFei31A8gSCOy1FJKiYkRB8mXndxdc8I9Mckuc7GBliUADxAg4kM3eIShT0NtNZacdCyGmu5NYr4rIo7H909oe5KlOvL62np42OPPbb1ZDwyFiPjIAYHI9MxAkvx71gHODpiBTHgWurgTgTYIzMkOosxeBIDSVGmMQZUCmsWxcBPlPrMD7rEGnnnn39+p88dpe3zQerotEanNjqrkd0e71XsJ/YXgaGY2RolkmOguau1iyKTKoL7+cHl+A7HYGusUxi3x3sUWffx3DFjNAbDCuJ17bnnnh1mJHQkAvoFcWLWm9fOgr7okEMOWWRCWfwexO9nDKbG3338dsVvTylBpkYWJ9vxegqivHBHkwzy5ecj2LDxxhvXpI1Az8UAW7HiicAF0aeK34V8ICnWiI7SlRGQjAHMCLZFfzKyL6OsdX4QNjLAS1lSK7Kk878rUVkrgrax/yiXGYG86IfFgG/0lSIjKJ9tFYH0/FqkXYm2R9nOEOtnxu949Fnj9UR/Mv4dEz4jGJgXry9f/aNYPL54maoYFI32xlrblRYZ1jF5Kl/6NSYrXHvttVm/Mvrc8fkU+sbR9nwlrej7x8SHrjL7o6R3VCcoiPc+AuDxmcTzPv7449l+4vgYffLo5+ero0Vp1QhIVkJMAIgge0ygKIggZHxeMWheLTH4HNWx8qLEbFSwivcg1lGNvkAEgeP7Fe9JZLPHMbIzce617777tulDRL8jMp3j/CL+tgrfyfiOxxrweVEtIkrudyXO5fJZ0TGB5pRTTsnOC+N7Ep9lfHbR7sIknbivq+Vo6vWdzIuJCIWJ0PF7Fb8XEbT+29/+1rqf2H8tFWcLxmfc3bV0q33+DlRHjHEVT9yLyozdWf6lkcRvavE4ZQSHe4OYFJmfiBaT26JvF8esOEZGxZZCvzYmd0WfM46xMcmrnCz/2K5QpSbOk+O7EPuIMdG4xBhjLKGTD3ZH/yHaMmnSpPSb3/wmu23vvffOjstxDIi2xb/j+BbZ+HmRqV5KhdBCrCGOm/EZRpZ79E/jPCD6MPG6I0EwlhSNzPq8yJyP4w2l6d1/5QB9UHHmbGQS9KYAdhyoiwckYtZ1XxVB2fwgWgwwVWtNueJOVGTm9EQhwykG0WIgImY45sWM3OjwxSBZ/iQ+Bpdi9m5nooMZHbJYmz0mJHS0xl1kPkSHLyaexFp1Bd/97nc7LelUvP/IrIoTmI5m20Z2eQwO3njjjdnnFeswduQb3/hG1tksiHJP8d5H8K14IDUGTyILIQbBYpZqQXSkC2u6l6r4/S/OFAHqK47HUWa2vUHTGEiO38oIrkTQJYI4cUIbvycxSac3LsWSL0MfgwT5mfcFcdJ/2223tbsN0PgiU7c4q6e94HD8hkWwJx98j3Umo1R2BECLs1fjN/Dggw/OBvQ++MEPtt4eA5cdBfhD9Kfywd3oP8ZtsTZnZK8Wi2BdPGf0YwuD2NHW4sHIzsTvdwRJo48X+/ra176W/d7nRQZvTPzcdddd29weEwM6KtsZQc04BuRFIDEGVKPdcYyIQFn0TWOiaCWCwvms6ei3xgBtfG7Fn3F8XnvssUd2f7y2ghjk7WwZgvhuHH744W2yvePYEBMm4vUUBxLiXCL6+TF4m1/nNiY89HSiWgR3I+j9+9//vs2gf3yG+e9cNc798uXEYxA+JgHHBPQYoC7OFov3JLLUovRsBIBjAkNHYuJyPjsyzl/ibyiqDRR/htHXiL+//fbbr83t8Z3rLNs5+ir5pa6ivdFPieoEMZkwL9od389C4KDcUua1+E4WK2Tpx29GTB6MgH9enKvHMmm1EoGX4optPSk/X83zd6B88Rsdx7jiSywdGUkjcQyMY1VkWueXKollRqO/UY44bra3r1Iu3V3Soyvx+vLViuJ3p7i6UqPK/ybGsTDGsaPvWjzmVzjWR98tJlHGuW8EyPO/sZ2JCW0xkT3elzgexPh4ftvYd/Rbo3+YF32LwtI9Ud0pxiGKKybEGGqMO+Yr0EVfMybkdSYSmGIMNiZvxUTJSCgq7r+EGPfYdttts2NIfhmh6PtWu6pTXyIAD9Bgitdg6yhw2aiTB/IZcyE6FtUKSNdbzDIvzgSImYBRBq4aist9VmI/sY5lV2sGRkcxBgjzuirvGYNCxYOxnYkyRzFYkO+k5te772yiSgw8RDnUUsVgVkclpGLSwI9+9KPW6zFoEx3zrkpORWc11h/Nl9uMDKriahadyZd6DNbrg8YTM95jsL/477VYBIBiQOZ73/teNgkpfq+jZG9xud5GFr/j+SVV2is1d80117ROLoggQ2R1Ab1HDCgWV/Zobz3vGESOTJuCGICM9ZW7ygCKMvERwMsHZmOSZUdicK9Q0ju2iTW8S6kEFucf+WB3TJyMAedyRJZxZ9Wl4pymuK8Xwb4YEO0sW7WjichxjIj1OqO0aJQ6jUmkEeC/8MILO10GoKMJ0PmgagQYY2JCPvO8PVECNT6fmDBRyucTffWoFFUQA7ylBDO33HLL7LUWRFA23svuiuoEcYzKT+aICbERBK9m6ds43kWwIS8GrmMScCni76Wjc5bIms9PdIvvf/zdxXejI/GY+AzyVbTi84nS+B0pPoeLzPyuJstHsPcTn/hEasTvZHsiwF/ppQ66K1/NI0SQI/95NdL5O1C++P2P38fiS0zIiiB7/K3mE5Ri7KpQQafUAG6+z9Hevkq5FE8E6qkI8sZvc/Hvc2T5d3We3CjyY4pxbC6e0NSZeGw5WfBRnSfOpTs7BsZE9nylnOgLxsTXGE/vqqx8HMvzfe2YWNeZmBQZS26WI8YyovJVgbXgSycAD9DgAfhhw4aVtX107kqZAVmJDlh0umIWZ5TliQGmyDoozuA46KCDepyp3YhicCMGzPLBlBggyc8KrLTiEj+x9k9PnX322SU9Lk4c8tlAMVsynyVeCTFYkj8J6ax8aD4Dq6uBnHJExn4+UzVmnbY3E7Q9Sy21VJvSqtFhjiyOUhV/nko6QWPaeuuts3J1Z5111iJr3HY2YBID2JGVF2Xneot8RnsEBor7KPmB5ZgdX60JaED1RMArr72M7sgmzfd38utYdiWCTbG0T0EE1duboBj9ynxFjQiq5zNhuxLnIvkKJRHEK1WUBC3Obm9PTC4t7ut3lmUU7YkS/XE+1JUISt90001ZRlxUP4tsueKqYh3Jfz6FQFupA8NRFSoyrAviHLG9tb7j9z8/ESsC4NF3L1VkleUzrMv5fPLi/GCzzTZrM1E1Mr+iGlZP12jtSnw++XPdGFCP0u2VECVj86XWo9RsKWX04zsW5yvFk0naK9sef2MRGM9PkCk1A7N4H43wnezIaaedlhpBfAbFkxd7kv3eiOfvQOliXDQqgUS1j3KCt42S2R99tJg48M1vfjMLFMfkg/yxJvoF5R4r6qlaY4rticSt4rXe2xPVN/Pie1JKPyOO5/nS/1F9paMKTT2xzz77tOm3WjazNALwAA2mODuulIN08WznUmZA5g+cpYgyk3Hwz19ihl0M2sXMuciIKD7Ax6BZrPPX10QnMwZ6osx4fvAjMgHzs/UrLV+yKvR0kCk6aO2V82xP7Kt4YDLKHlZSfNfzA3OdlSjNz6qN8k+dPbYchfWVCoM/MfhZbnZsXn6d+q4Uf57FnzfQOCIAFcGeKMcbA8NxYhwB6OKyxcViADmy9PLlaxtZlLYtDBBFecEo51oQx8D80ijKz0PvVNz/yJcSDbE2Zazfns++bK88Z6n9ozhfaK/fVpytU1xauyuRyZ8ffCynD1bO71f0h/PvWUy2jHUzO8syiuBqTFiOiQilDPLGuUZM1oqytdG2/CBxsagYkM80jqBtuZOfS+m/RnZ5vh3lfj7xuvPZ3xFIL1Q7KFWUIo++eX4yWJR2jyys7q6pXY7i6lxRMrZSg/bF51WlTNooiHP7/AS46GsUL2tX+AzzYgJFqe1fb731soplpajVd7KjyTSxxFkjuOuuu9pM2IhlKXpaGbDRzt+B0kWlkyOPPDKNHj26yxLhjZjZH+ewcdyIbOt4LQVx/I1KoNFvyS830+jyY4ox4bESywG1J86lS5nkGYorMcWxt9RldfLVVeI4/Pzzz6dKK14et1LjsH2dADxAgykO4HY2oNOoooMRa9hFmd5yypD3FlHiMkoC5kW58cjGqKZ8ycfQ085tZHH25PH5wdjOxHqHMYEjBh2j8xjLKsQARPGEjrjky57GmsMdyZdBjCyqOBmI0vHF71E5XnvttTblVaNEYPFall2Jk6m8crIMIqDX2//2ob+J362YbBZrp0ZmQKzdGgMSkfEUWU7tBamiykZM4ipeVqQRxW9a/rc/n/Ge/3f0XXrLen9A55N/iyv/3HvvvW0ynEopCd+d/lFxgK2n+4ky5e1lAfe0Txz9tQjy5fuhkSnWlQhCRgA51ueOgHGsjx3LfHSVBffTn/40Czq/9dZb7d7/2GOPtZm02Vs+n7lz52ZLTpUqyqrHAHZhEkD00WPd0wgE1ErxexBLzFRCfLb5id0RFN98881L3j7ei+LS9vlM947O3Xp6LtiRWn0n21NK1YBaKS7PG5OXipf8aNTzd6D0iTbR12jvEhMOY9nKyBrfcccd2xz/vvjFL2bnj+WIipsd7aury84771yFV59SU1NT1k+M9cB7U/C9eEwxjllxHI3lJ2O99koHrWOt9lIUHyNiTLJUxdvG96wU8XqjitBXv/rVbMmgmNAXSQXRtyger42lFfI6G7Plf6o/RRSAshQfmHtLFmwcnKMMURyQo6xSrcvOv/nmm1nHrxTRMexusDxKDhevnxYd55hwUGulDip2pLN1Lkt5fFfBoxh8Ofzww0sun1ksAlkdibWlorpCoRxhlD+K0pax3l8MUkbnOT7jyNQoNbMjgmb59zSy4XtaGqy4XHM1P0+g/uI3IwIqcYnswMgkjQHYWCM2/3sQAa9YRzbKxDa6yL6MwaXCwG38Vq6++upt1j2OoEjxJCKgd+iqwlJx4OuYY47JLpXuHxXvJ18VqTsiMB6Dj10t5xXLH5Wz7mehT1z4XSz0iWPQshQxETUmYcUlRBv/9re/Zf3lqALQXuA0ssUjGy3Wh+/qfYugdFyq/flUItAZ+1l55ZVLKqWdX3s9zuXiGFRuQCEm+sbE4FIz44ozjaPqTb5yV36t1p546aWX2lSeiO9jTFYuR2TJxWSNgvYy+YrP3co9Fyx17fJafSdLyc5rT1RPi4kwpVh77bW7texbBN5iyYJKl5+v9vk7UDlxnIilSuISZeejkti+++7bugxPLGkTE/pick5v1dzcnI0BR4JSudUjeyqOyfnjcmcKn0Pe0UcfnVUxLSTxxHPF5xRVXuM1xQSmmGgfv6PlJuYUT1IoVfH5dE+27ax6UmH8MSoxRCWhGE+t9Jgt/yMAD9DgAfgo+1iOjmagRcm54pnx5YhAQnF5xgg0RKcyBrZilnq55fIrKQYu8jMYu5o5ml83sFSXXnppFkTJi3XzKrX2X7kdqph00BPlzsAvfnxnna1Y7zICMsVlTMvRUaZP4e8k9vHZz362TdZ8dJ5vueWW7FIYRN5iiy2yzMzddtttkXVOe/K3VopyJtAUd5Dr+fcEVEYMoMektDg+xQl8HKsKYhAmMvpqUTa3J+K3MyZTFapyRHZ/LD2TP1EvBJKA3iUG34rPHfIlOWvZP6rWfroKwMf95f4Ol9Mn7kpkGcUgb1xOOumkLHgZa4UW+rIFl1xySVaFa5VVVulTn08p8sH3+Kyiylp3JlPH+rVReaAUcVzLZzDHRImoYJOfSFGpNXyjCldePHe5ircpfs72vqc9PRfsSD2/K10tBRRi6bx8FZ/OXHHFFd0KnEcWZb6a2fLLL5922GGH1FPVPH8HqivKt8fyXd/5zndab4vJdZEd32hrwsckw+IKGjGxKILUkagSSUmFc8H4fY5xuVjiIl8hqNrinDQqbZYikqaKx22jOmdMHIjPJZ8tHr+TUbUgLiEq2sVYepwTx8S/cjP9e1IVtifbdpbgEyXqo59z9dVXp57obMyW/1GCHqDBRNZcXnRmuhMsrrSYLRgzGvOXWO82BuFj9n9fDxZGlkVxlnt0WCZPnlyzNhQHj4tLhpar3GzF4s84OuDtiTUHo3OaD77HCcVHP/rRdMQRR2TZOzEoEWWOYgAtf4nBiVJFOcMoGR8zVzsq6RRtjGyiyI6PSSLR6e6opFQ1BiTKWduy+P3sasAY6D0iIysGfIv/5vMlZxtVTGTKl5ePE/UYkM5Paot1ioHeJ8q0Fy95ExUu6tE/qlc/rDvVO0rtE3dHnFf98pe/TMcee2yb2yP4WxgM7k+fT/GyCPE+RPC01pWjis+7iitF9ETx96c759XF27R3nli8n56eC3aknt+VUiuf1br8/D777FORCZfVOn8HaiMSePKistj999+feoM47q277rrZJMGodhmVV/LJJHvuuWevW0YxJkZFZZyYMN/RcT0mtEVFk89//vPZeW+MZ/b26pWnnXbaIsH3mMAWEyliUkNkxsdrjskW+fHac889t25t7s0aO90CoB/aZJNNslKIhbJEYdq0aYtkO1A7MQgWwfb8iX9kd19++eU1nalaXKIxMr97Uvqw3PXSizvTHXVQo0OenwkZJSpjoKyU8p7lvp8ReI+g1umnn55VeYiZujHzNv5migPtMRAVnclYp/n2229fZACj+HrMco1yTD3R3vrPHSkuhVi89iHQu0UQ+6CDDmqTQRfr33ZnbdRaiwo4hdK2UdY2X9o2KuQ0WtYGUJr21gPeYIMNOu0fTZw4sc1apt1RXAa0sJ98BlJMoOxpwOr9739/xfvD5fSJeyLWNr/55puzia35NciLA/PFn8/ee++dHWt6orgKQnv7iYlYsU5oT+QH7zszZcqUNGHChNbqX7F8S0z0jQHinpSFLcf73ve+qgUyi78/3QlgFG9T3N72grHx3W/vcaXuoyO1+k42qijzHn+rlS4/X83zd6A2ol8SfaB86fQ//elPWaW03iSWCYqEmljysbBESPz2RcLLOeeck3qT6MtEtdPvfe972VhhjCvGskCPPPJIm3H5Qsn9mEQRY44xKTLG7nubWMbz7LPPbnNbVHqNsc+ujhHF7welEYAHaDBxwIuBrwggFsSss8gopmMxQaEasxCjAxYzOfMBk1gPKDLia93ZKp6E8fzzz/fo+TparqDUcoLtlXOPwahf//rXrdcjo/13v/tdyYHo9solllrmefvtt88uhRm4sYZm/O3EZ5UPbscJTpT6Kl6LsLh0Y5R7quU6VsUBeJNuoG+Jge/4ncmXoe9O4KcePv7xj2eDE+0dd4qXpwF6jyi9mbfGGmssElQt7h/FesjV6B/FfvIB+A033LDH68CXIqqNxaTNcjJnS+kT91ScZ8TErbPOOqvNpK1ixZ9PtKVan09erI9eiXXgS/GpT30qO7/YaaedWo+bMbk3gvBRgrbUiRoRBO1uIDQyw2I/hXPCOI+Kc89KTEArPk/qTgn34vO69s69ir+nsU05AfhS21Wr72RPstOLM9QrqbhCw0YbbZTWWWedijx3Nc7fgdqKvk0+AP/cc8+l3igShCLY/sUvfrH1tgsuuCB95StfqclYVgT7K7kcaJyrf+5zn8suIfqk9957b9b/uPbaa9uMVd54443pu9/9bjrmmGNSb0wwyy9/Gdn/+aV+OlOYbEF5lKAHaECFA37Bdddd12ZAitqIQG18Fvls7lhPPLJRIuBba1HuqbhcVU9E+fZyRJmp4pLKxf72t7+1KT0fGQ+lBt///e9/V2wNoViXKTLY44QglnA48cQT29z/ox/9aJG/qeLXE+2pdRnYzj5voPcrzkTqzjqv9RAZhvvuu2+7VXti/Tyg9yms45kXWcbFatU/qlc/LLJ5ivtglegTV0LxAHZ7k7b6+ueTnwgWk3rzAeMYEI+1WztaXqrSYoJK/nj+5JNPVuR5Y1JL/twyvo/586lSRKZeXpTJ7aryRLnngrF2cW/4rtRTBN5jUkg1st+rdf4O1FbxMas3r6Mdv2/5cas4dkV5874gJt59+tOfTt///vezSRIHHnhgm/vPO++8XlmKPhKV8oqXWu1MlOunfALwAA3osMMOa1O6LU7wzz///Lq2qb+JQHKU18wPdEWZ4FtvvTUL7tZDzJ7v7IS6XFE2qSePby/rJcoZ5X3oQx8q+fnvvPPOVA2R0RQl6jfffPM2Jz1/+ctf2jwuMr7yA2v/+te/ajobOf95RnZCvi1A7xeTporXZO1NS03EUiyl3Ab0DlENKL+8Uky0iSyYYjGhsRb9tVrtp6d94jg3yFcqi0z14j56LSdtRX88f9543333tZZq7yufT34idCwjNWzYsDZZaLE0WC0CGLH/vOIJLN0Vwff80g8RwIisu1JFACBK5uZ97GMfW+RxG2+8cUXPBTtSq+9kI4qyxVGGOf/ZxrrBlVKN83egtvJLeYVaVPuplug7fvvb325zWyxb9uyzz6a+JLLjf/jDH7aZGBlV7XrjBLNGHLPt6wTgARrQyJEjF1knLQKIpc46p2eeeOKJrMx8lKQs+PCHP5xlXcQsyHpZffXV26x/F5MEeuLBBx/MXmsporR8DHDlbbXVVos8rngGaKnZG7HdxRdfnKpps80267KEX5S4zLvoootSrTrB+dLOxQN8QP1F5YyerPkaZery4gS+nBPeehs7dmz2+mMSQeFy8MEH17tZQDdE1k5UdCpeTqK9yX8f+MAHsn5wwfTp07P12SutuA8WA521ymwuzljtTPSH88eCCL4Xr61dKX/961+7nLQVAb7IDs8H7WN99kqLEuL5Uu8///nPu1UqvacisPyHP/whLbvssq23xTq0O++8c9WDvMXf0Th3qdR3tPi8qpwS6b///e/bTBqOjPo111xzkcdtvfXWba7HZ1hq+yPD/uGHHy7psbX6Tjai4s8tlk3If1d7qhrn70DtPPTQQ9k64nnt/V73Jp/97GfTeuut13o9jiulljTvTaIP9NGPfrRHy4I0gu6O2UYf4P77769Sq/o2AXiABhVr2cSaOvmDYnRsyi2RSHlixvonPvGJNh2pGIyMgY2YGFFvhTXOC+tAxiBoTxx77LElPe6MM85oU7I9Br/WWmutRR73/ve/v831UrM3YgCr1EGd7iplbcQjjjiizeDihRde2OOJDqVmS+TFBBCgscREuCjdGevxzpo1q+zg+2WXXdbmtv322y/1NhFkWnrppVsvkfUA9B6xfvU3vvGNdOSRRy7Sfzv77LM73O7oo49uc33ixIltJqpWQgSy81nWEVD85je/mWohAt3Fgar2RID35JNPbnNb8aTpgqi01JOByujjxzqdpfQPiz+faGNxhl1PLb/88m2OWxFUjXVe6yG+K3fddVdqampqvS0mSn/mM59pt0x/pcTSZPnst1jGoVLrz8Yauvn15K+55pos2FrKEgrFa9B+6Utfavex48aNaxM8iHPJWK+3FMXfsXIfX43vZKOJv4kbbrihauXnq3X+DtQu8Pmtb32rzW1xLrXDDjuk3iyOXcX9tZiM1FvXtu/pmGKj686YbfQ1yilVT1tGTAAaVAR7f/GLX2TlswuijE9k8cZs9XLXmvnnP/9ZhVb2LS+++GKW3RH/z2eaRJZFcSelXqIsfl4MPvVElNTvanZqZEjFWurFA68dDYjl1zC86aabsrKDXbWheCC4MzGQsM8++2Szh0sV68DnB0TiRGf99ddvd23EGAAreOONN7LBvHIHUKM0U3tlXDtSXDYy1poCKicGse+4445uXYpPuo8//vgsIzQmxV1++eXZxLj2jsmRIR4BnS233DIr9ZwXE+xKHUAF6KkIHMckoFin83vf+16b+6JUdPSROiuBGv2utddeu815RQwY5/vMXYmMqKuuuqrTQH+sG5qf2DNp0qSstGk55z1RUSiCf8XZ412Jih6dra8c5fojAJ0vLx3vWUflpWN98E033TR7n6JfWM5riEHrOMZEPzS/PFHc1p44zuSD85FdF5N2y5m4Ha9vypQpnR6bYoA9X1o8zlUPPfTQstYrf/XVV7MJbZG13hPxXY7+c/4cLc7ZIku9J9VqOhPLDRQHGb7zne8sUuGmI/Ed6OjcLTIg45wj/3lEaf18haz2ni+C7fn132Oi3P/93/91uE3xOdyJJ57Y5flkBIxiMno5avWdbCTF1THiu1lcNaESKn3+DtRmgk78Xv/6179epPpQe8vL9DZxvMr3E6NfEJPWG1X00WLp13JKyEe/Mj9uF/2yGD/sbaJvWnyMz/c32wu+x2RT2e/d978ULwAaziabbJKVX4yAYGGNxhi02HvvvbPOTNweAeP2ZjLHCXkEHW+77bZ09dVXpz/96U91eAW9R2RLxKBAZDLkB1mOO+64bJCx3AkMsd74kCFDKt7OGJiJMviF2ewxKNFRlkNXxowZk03qiIGXGHCMzIR8OeSZM2emyZMnZ4NK0ekqiPdpzz33bPc5Y9AnOt/XXnttdj22i4HHGACIk4v8exJrrEcJ1EsvvTT7fscgZjy+q3KW8dif/exn2WX8+PFpt912y6oWRHnUfPA/xPt0/fXXZ4Nl+UytKAfY0aSKaNMDDzzQGuCP9yEGkfbdd99skDH2mc+SDzHYEhn8UZI1Bl9ired4f0sRf6u33HJLm7/7yLIFKqc4AF6O9oImkUUawYNCACGyweM3LCbPFX7HItOrvW0jYy/Wja1WyWKgf00uit+j4kD37Nmzs0ucC/z5z3/O1ixvLzM4spoj+F68TE+x6BNH/yayZwv9qRiIi77XV7/61SxA314J1VhiJwYs47cyAkIRhNt///073E+0I7I2Y6JTQfRPo58Uv+MR0CvONorf3Oin//GPf8zaGMHueE+KJ612JAZQo2/52muvZYOSsf9oY37ZqXitsf/iSaXRT37f+97X6fNHZnZcol8YwfooAx6ZqMXLWsXxIgaEo397/vnnLxJEjokInZWyjjL60UctZJxFXzQmxsbEggMPPDALWOczrEO85vgORUAg3rvYtrMS1THQHBPP4ly0IM5VY0A63p+YIBDfqeLXFedX8d7FdyDeixjorURJ8sjojjWuo9x5ZHOH+B7EuUr0yfNrxVdKnH/HayhM7I3XF6893sM4b4z3b4kllmh9fHy34nONagbxGcVkmPi7bM8PfvCDrP3xtxviPC3Who9JK3Hule83xN91BKaLq2ide+65bZYsK7bXXntln2FhguFbb72VBYlPOOGELLstX1UgSp2fdNJJrdUhIvu/o7bX6zvZyOXn43ex+Jyxp6px/g5Uth9WEP2u6AfF2FL8nsXvW15M6I5JXKWK40fx5PByRJn4/G98NbLg8/2DONbEb1Vnx6R6iWPfJZdckvVhYuw2xhSjClOMrUefNy/6rrGufVS8yf+uRl+xePyxN5gwYUKWABXJAiEqfsZrj75nvkpOfK9jYmN8hoWKPPH+RJ+G8gjAAzS4KFsWA03x/8LJeIiZ7l/72tdaB/6jIxUzJ2MQIA6kMVs+Zlm2Jx4XJ9P8T5QT/vvf/16xMjuRHZMvUVgpEcCOzuGPf/zj7Hp0wOPz7mrwrz0xmBMdr8heioB5XCLjPwbO4uQgBssKEz/yJ/3Rke5MZC/FoElhkkD8PwLXX//617PB2RiUisGBfEZHdHJj0CJmoZaznmQMKMclBryi87vSSitlfy/xfPE8MUiU7ySH+FuJ0vIdWXLJJbOB3shCf+yxx1o7n9G+uMTgV2SvxqBenFTFexWDfuVWpci/hvx7EYF+oPFEwCT+1tsbZIlASVzyk7jaE0GCmHTU3jrLALWcXBSBuJh0WGqVpwjyRAA1JloWBpHj/xEYjkucX8RzRT8p+n5RMaR4ndNSRJ8u+uXRtoIY+ItB3ciOj75qYVmoODeKPmVPyo5Hfy76rjFRNPrUcX4VGfQRbI5s7wgAtrfsSARi8wPNXYmgWQy0xyVeR0zaivcszuNivzFpqzAYWizORw4//PBOnz+eL/rfMfGgEPCM9yUGVOMSrzMG++OcIY5XMak73rvufG/i3CHeo8J5QkyAiMpPcYk+cryuCDzG5/PSSy91+LoqIc4tIggdx9d4jwsTJmKS+u23316V8rBxLhTf7Qj+F0QWeVziOxPvQUzsKHyu+ckUnU3QjfOYmDgff2MRHAjxdxTftfj8Y4JunKd09J2MDLXOst8LYvJDBBwK71dkKkZgobDUTrS9+FwtvjuRABCff6lq9Z1sBPFeFlc0q0b5+WqdvwO17YfF731UFimn0mYE8yPxpLuiD7fzzjunatljjz2yY0lMtgpxHIvfrPi9b1Txexl9iMJktjjGxnGp0HeIY217k+o/+MEPZn3H3ij60JH1nl++JpZNirGOOJbEMSUme8RxLb+MydixY7N+QCzHQ3kE4AF6gTjAPfroo9lAx3XXXbfIwb8w8J8vidie6ETECXzMYIsTa3qnCFIXAvDRqY3vRL5seqmicxUVEiKbqFBCNDqXHa3PFwGjeHycLHRm9dVXz7LOY/AoP+AUnbj4Hrc3qSAC2z1d+yoGj7oKfkWnMTJQYmCsM/EaY/AuBrFiDcb831xMbCmlfGJ0XEsRs2kLYgBKAB4aUyz/EoPDkckZk5/iRL2U9UwjGBXVS2KWfG9f4w/o3SKDOvpnEWSO7PVyRYZMZLRH4Lm4xHsECovXxmwvQ6qrPliIMvmxVFBM3sxPQI6B0phc2VUWbvSnyjnXidLyMbAdg5HR54v+dWcZPtHvjqypzsRE3OhPtldCPF5HBKbj0tXriAHe+LyKM4Xbs84662STFeL1RJ89LyoX5KtB9aT/GgHAKDUbWczFAdMItHa17mtMxu1syYNyxWSJQhB++vTprRNc43oEOCpd3jeqF0Rw/ytf+UoW2MyfJ0SAuRB86I4IVsdnF4GMfJA9qgZERnp7YuJxfHe7KkteEN/LyGqLc8DC+xVigmFUKCsW2YuR9d/V33c9v5P1Ftn++e9BZPp35ze2XufvQG3E8S8m50RguhpVWuopJhfGWHNMaCyI5Y+iqlGjLOnZlTjWdlWWPipWxmSG7iRBNYqILcTxPxID8qIvHJf2juVRwa+ckv3ktADQq/zjH/9o+cY3vtGy2mqrxRlel5fll1++Zeedd275+c9/3vLmm2+WvJ/999+/zfOcfPLJLbVU/Dqefvrpqu4vnr+U97PUS7Xbu9lmm7Xu62Mf+1hJ28RnmG/jXXfdld0+a9asli996UstSyyxRLuvZcSIES3HHntsy4IFC8pq45NPPtmy0047dfgeDRo0qGW33XZreeqpp1q3GTNmTOv98e+OPPTQQy2nnHJK9j501O78Zd1112254IILWhYuXNhSrkceeaRl7733bhk+fHiX+xk7dmzL17/+9Zb77ruvpOeOv8lll122dfuvfe1rZbcPqJ/4/fzjH//YctVVV7Wcd955LaeddlrL2Wef3XLxxRe3XH/99dnv4LvvvlvvZgJ9RL6f1NFl8cUXb1lmmWVaVl111ZZNNtmk5Qtf+ELLpEmTWv70pz91qx/UkVtuuaXl4x//eLa/ztqz2GKLZe349re/3TJ9+vSy9jF79uzsd3XNNdfs8nVHfzX6lT/5yU9a5s+f3+nzdtTfnDp1astHPvKRDvexzjrrZK+7VO+9917Ln//855ZvfvObLVtssUVJfda4rL322i1nnnlmy8yZM1u665577sn64UOHDu10XwMGDGjZYIMNWo477riWRx99tKx9vPHGG1n/OvrZ8Tyd7WfppZdu2XHHHbPj46uvvtrp82611VZtti3V888/3/KhD31okffypZdeaqmWOCfZZZddunyf43sW5wjRLyjF3LlzW0444YSWFVdcscPnHDJkSHaeX+7nVhB/J3GOF3877T1/fF8PPvjglubm5uzxce7Yk/GBanwnr7jiijbbxvV6WGONNdq048ILL6zI89b6/B3oeT8s3x9ramrKjkP77LNPyw9+8IOWV155peR9Ff/m9vRy8803l/y6Cr815XrnnXcW+T088sgju9yu+Lc8+gHVFOfn0S+O38mNNtooG5vs6v3bdNNNsz5m9O260t3X0pPjbEfHi87E6+kstjBq1KisH16IJZTbvlLHdxvpeF4NA+I/+YA8AL1HZBxERnGUhomSY5EBHLPwItM9ysrELLVS16Gmd7n11luzdcwLYt2eWCOwJ6K8UKw7GJkPkeUd2SKRURJrVQ4ePLhH39N43sgAiqyQyBqJ2fixzmYlKjFEllJkhMRszNhXZN1HllDsJ7KPIoMqykj1VGQrxfscZTYjAyTerygzGa8hsv5jHcpy19SKbInCWqjxHkfGjPXfAYDeIvp2sR51ZD3H8j+RPRRl1aMfGWXrY73I/NrV3RVLgETWfWQFx34i0yr6etHHi31EXyxuK0X0Dwvlt+NcqTijPvqU8Zpin9GnXGGFFdKGG26YZX33xNtvv509d/Qlo88afcm4Ld6vyIQr9FsrWaksnj/Wn41KadF/jT5+fB5xvhjl26P/WoksvCjJHiVMI6s/Pp/oN8fnE1lv8flEudaenE80ujgfiepZ8b2K96JwXh7Z25EJHd/P7nr44YezpdLiux/7ifONqCQRZeTjXKSnoq1RTj8qicV3JL4f8XltueWWVcnwq9V3sj+o1vk7QH8Sv59xnI2s8MgAj+uxlE4ci+I3NcZayx3r6y0iNBz9jKhWE8fkuB5ViqLvMn78+KzKDj0jAA8AvdRHPvKR1vKfUQr0Zz/7Wb2bRJnWXXfd1nXmY93M4hJQAABUVlcBeAAAgJ4qbXowANBwzjjjjNZ/x5rrkUVA7/HrX/+6Nfg+ZMiQ9M1vfrPeTQIAAAAAoIcE4AGgl/rEJz7RWob+nXfeSSeffHK9m0SJogDRt771rdbrRx11VFZKEgAAAACA3k0AHgB6sfPOOy8tscQS2b+vueaa1oxqGtsvfvGL9NBDD2X/XmmlldLxxx9f7yYBAAAAAFAB1oAHAAAAoF+wBjwAAFBtMuABAAAAAAAAoAIE4AEAAAAAAACgAgTgAQAAAAAAAKACrAEPAAAAAAAAABUgAx4AAAAAAAAAKkAAHgAAAAAAAAAqQAAeAAAAAAAAACpAAB4AAAAAAAAAKkAAHgAAAAAAAAAqQAAeAAAAAAAAACpAAB4AAAAAAAAAKkAAHgAAAAAAAAAqQAAeAAAAAAAAACpAAB4AAAAAAAAAKkAAHgAAAAAAAAAqQAAeAAAAAAAAACpAAB4AAAAAAAAAKkAAHgAAAAAAAAAqQAAeAAAAAAAAACpAAB4AAAAAAAAAKkAAHgAAAAAAAAAqQAAeAAAAAAAAACpAAB4AAAAAAAAAKkAAHgAAAAAAAAAqQAAeAAAAAAAAACpAAB4AAAAAAAAAKkAAHgAAAAAAAAAqQAAeAAAAAAAAACpAAB4AAAAAAAAAKkAAHgAAAAAAAAAqQAAeAAAAAAAAACpAAB4AAAAAAAAAKkAAHgAAAAAAAAAqQAAeAAAAAAAAACpAAB4AAAAAAAAAKkAAHgAAAAAAAAAqQAAeAAAAAAAAACpAAB4AAAAAAAAAKkAAHgAAAAAAAAAqQAAeAAAAAAAAACpAAB4AAAAAAAAAKkAAHgAAAAAAAAAqQAAeAAAAAAAAACpAAB4AAAAAAAAAKkAAHgAAAAAAAAAqQAAeAAAAAAAAACpAAB4AAAAAAAAAKkAAHgAAAAAAAAAqQAAeAAAAAAAAACpAAB4AAAAAAAAAKkAAHgAAAAAAAAAqQAAeAAAAAAAAACpAAB4AAAAAAAAAKkAAHgAAAAAAAAAqQAAeAAAAAAAAACpAAB4AAAAAAAAAKkAAHgAAAAAAAAAqQAAeAAAAAAAAACpAAB4AAAAAAAAAKkAAHgDqZOutt04DBgxIBxxwQL2bAgAAAAAAVIAAPAAAAAAAAABUgAA8AAAAAAAAAFSAADwAVFFzc3NdtwcAAAAAAGpHAB4AquTVV19NY8eOTYcddliaO3duWdu+9NJLabfddkubbLJJWrhwYdXaCAAAAAAAVI4APABUyaRJk7Ig/CWXXJLWWmutNGXKlC63aWlpSZdddln2+BtvvDFNnz49uw4AAAAAADQ+AXgAqJLjjjsuTZw4MQ0ePDi9+OKLacKECVlW+8yZM9t9/D//+c+0zTbbpEMOOSTNnj07DR06NJ166qnpoIMOqnnbAQAAAACA8gnAA0CVDB8+PJ133nnp8ccfTzvttFN2W2S1jxs3Lstqj2z38Pbbb6czzzwzrbfeeunuu+9OAwYMSPvtt1966qmn0kknnZSWXHLJOr8SAAAAAACgFANaCqP/AEBV3XHHHenII49Mjz32WHY9Au1xGC78P2y22WZZ0H7jjTeuc2sBAAAAAIByyYAHgBrZbrvt0kMPPZQuvvji1NTU1Bp0j/+PGTMmXXvttenee+8VfAcAAAAAgF5KAB4Aaiiy3QcObP/w29HtAAAAAABA76AEPQDUyJ133pmOOOKI9Oijj3ZYgn7zzTdPkydPThtttFGdWwsAAAAAAJRLqh0AVNn06dPThAkT0rbbbpsF34cNG5YuvfTStMUWW2T377333un0009PSyyxRGsJ+gMPPDDNnDmz3k0HAAAAAADKIAAPAFUyd+7cdPTRR6dx48alKVOmZLftsssu6cknn0yHHHJIlvkeBg8enE488cQsOL/llltm2fBXXnllWnPNNdOZZ56Z3nzzzTq/EgAAAAAAoBQC8ABQJWeccUY699xz08KFC9OKK66Ybr755nTjjTemFVZYod3HR8B96tSpWXb88OHD07x587LA/OWXX17ztgMAAAAAAOUTgAeAKjnmmGPSiBEj0qGHHpplve+8885dbhNZ8ZEd/8QTT2TZ8quttlo6+OCDa9JeAAAAAACgZwb1cHsAoAMjR45MTz31VGpqaip728iSj2z55ubmtPjii1elfQAAAAAAQGXJgAeAKupO8L2S2wMAAAAAALUjAA8AAAAAAAAAFSAADwAAAAAAAAAVMKClpaWlEk8EAAAAAAAAAP2ZDHgAAAAAAAAAqAABeAAAAAAAAACoAAF4AAAAAAAAAKgAAXgAAAAAAAAAqAABeAAAAAAAAACoAAF4AAAAAAAAAKgAAXgAAAAAAAAAqAABeAAAAAAAAACoAAF4AAAAAAAAAKgAAXgAAAAAAAAAqAABeAAAAAAAAACoAAF4AAAAAAAAAKgAAXgAAAAAAAAAqAABeAAAAAAAAACoAAF4AAAAAAAAAKgAAXgAAAAAAAAAqAABeAAAAAAAAACoAAF4AAAAAAAAAKgAAXgAAAAAAAAAqAABeAAAAAAAAACoAAF4AAAAAAAAAKgAAXgAAAAAAAAAqAABeAAAAAAAAACoAAF4AAAAAAAAAKiAQZV4EqD63nzzzTR9+vTW66uvvnoaMmRIXdsEAAAAAAAA/I8APPQSEXz/8Ic/3Hr98ccfT2uvvXZd2wQAAAAAAAD8jxL0AAAAAAAAAFABAvAAAAAAAAAAUAEC8AAAAAAAAABQAQLwAAAAAAAA9Albb711GjBgQDrggAPq3RSgnxKABwAAAAAAAIAKEIAHAAAAAKpCFiIAAP2NADwAAAAAAAC9RnNzc123B+iMADwAAAAAAAC9wquvvprGjh2bDjvssDR37tyytn3ppZfSbrvtljbZZJO0cOHCqrUR6N8E4AEAAACAbpOFCEAtTZo0KQvCX3LJJWmttdZKU6ZM6XKblpaWdNlll2WPv/HGG9P06dOz6wDVIAAPAAAAAHSLLEQAau24445LEydOTIMHD04vvvhimjBhQnY8mTlzZruP/+c//5m22WabdMghh6TZs2enoUOHplNPPTUddNBBNW870D8IwAMAAAAA3SILEYBaGz58eDrvvPPS448/nnbaaafstjiejBs3LjuexHEmvP322+nMM89M6623Xrr77rvTgAED0n777ZeeeuqpdNJJJ6Ull1yyzq8E6KsG1bsBAAAAAEDvzUJ866230ve///3WLMRdd901XXjhhWmFFVZoNwsxMhAjEBIiC/GYY46RhQhA2dZcc810yy23pDvuuCMdeeSR6bHHHsuOMRFoD9dee21rMH6zzTbLgvYbb7xxnVsN9Acy4AEAAACAbpGFCEC9bbfddumhhx5KF198cWpqamo99sT/x4wZkwXi7733XsF3oGZkwAPQ57zzzjvptddeq3cz+owRI0akQYN0GQAAgI7JQgSgnuJ4M3Bg+zmnHd0OUC0DWgo9X6Ch/f3vf08f/vCHW6/HzPK11167rm2CRnT11Venww8/PM2ZM6feTekzhg0bli666KK077771rspAABAL/Duu+9m2e+R2d7c3Nx6e2QhnnXWWWmvvfaqa/sA6FvuvPPOdMQRR6RHH320NRgfoa/C/8Pmm2+eJk+enDbaaKM6txboD0z7AaBPZb4LvldevJ/xvsb7CwAA0BVZiADUwvTp09OECRPStttumwXfI4nk0ksvTVtssUV2/957751OP/30tMQSS7SWoD/wwAPTzJkz6910oI/T4wWgz4iy84Lv1RHvq7L+AABAKVmIG2ywQTr00EOz7PdCCfr4/7PPPpv23HPPLDDy4IMP1rupAPRSc+fOTUcffXQaN25cmjJlSnbbLrvskp588sk2y58MHjw4nXjiiVlwfsstt8yy4a+88sps2ZQzzzwzvfnmm3V+JUBfJQAPAAAAAPSILEQAauWMM85I5557blq4cGFaccUV080335xuvPHGtMIKK7T7+Ai4T506NTsuDR8+PM2bNy8LzF9++eU1bzvQPwyqdwMAoJqeeOKJtNxyy9W7Gb3OK6+8ks0iBgAA6CoL8bTTTksXXHBBFggpZCFedNFFWSDkZz/7WZssxN133z0dfPDB6Z577smyEG+44YZ0/PHHpyOPPDINGTKkzq8GgN7gmGOOSZdddlnaY4890qRJk9IyyyzT5TaRFR/Z8TvttFO21OLDDz+cHY8AqkEAHoA+LYLvTU1N9W4GAABAn85CDJGF+P3vfz/tvPPOHT6+kIUYgZNjjz02zZ49OwvMR8b8V77ylRq2HIDeauTIkempp57q1phfTA6LbPlYJmXxxRevSvsAlKAHAAAAALqdhThixIhszfdYe7ez4HtxFmJULIts+dVWW00WIgBl6WnCjYQdoJpkwAMAAAAA3SILEQAA2pIBDwAAAAB0myxEAAD4HxnwAAAAAAAA9AlTp06tdxOAfk4GPAAAAAAAAABUgAx4AAAAAKAqZCECANDfyIAHAAAAAAAAgAoQgAcAAAAAAACAChCABwAAAAAAgBrZeuut04ABA9IBBxxQ76YAVSAADwAAAAAAAAAVIAAPAAAAAPRbshABAKgkAXgAAAAAAACooObm5rpuD9SPADwAAAAAAABUyKuvvprGjh2bDjvssDR37tyytn3ppZfSbrvtljbZZJO0cOHCqrURqB4BeAAAAACgT5OFCEAtTZo0KQvCX3LJJWmttdZKU6ZM6XKblpaWdNlll2WPv/HGG9P06dOz60DvIwAPAAAAAPRZshABqLXjjjsuTZw4MQ0ePDi9+OKLacKECdnxZObMme0+/p///GfaZptt0iGHHJJmz56dhg4dmk499dR00EEH1bztQM8JwAMAAAAAfZYsRABqbfjw4em8885Ljz/+eNppp52y2+J4Mm7cuOx4EseZ8Pbbb6czzzwzrbfeeunuu+9OAwYMSPvtt1966qmn0kknnZSWXHLJOr8SoDsGdWsrAGhAI0eOTLNmzVrkNsrnvQQAAPpSFuJbb72Vvv/977dmIe66667pwgsvTCussEK7WYiRgRiBkBBZiMccc4wsRADKtuaaa6Zbbrkl3XHHHenII49Mjz32WHaMiUB7uPbaa1uD8ZtttlkWtN94443r3Gqgp2TAA9BnDBw4MDU1NbW5xG2Uz3sJAAD0FbIQAai37bbbLj300EPp4osvzsbZCsee+P+YMWOyQPy9994r+A59hAx4APqcd955J7322mv1bgYsYsSIEWnQIN0vAACoB1mIANRTHG86SnCR+AJ9y4CWQq8SaGh///vf04c//OHW6zFre+21165rm6ARXX311enwww9Pc+bMqXdTYBHDhg1LF110Udp3333r3RQAAOjX3n333Sz7PTLbm5ubW2+PLMSzzjor7bXXXnVtHwB9y5133pmOOOKI9Oijj7YG4yM8V/h/2HzzzdPkyZPTRhttVOfWAj1lSg0AfSrzXfCdRhbfzfiOxncVAACoH1mIANTC9OnT04QJE9K2226bBd8jOePSSy9NW2yxRXb/3nvvnU4//fS0xBJLtJagP/DAA9PMmTPr3XSgB/QmAegzouy84DuNLr6jlkgAAID6ZiFusMEG6dBDD82y3wsl6OP/zz77bNpzzz2zwMiDDz5Y76YC0EvNnTs3HX300WncuHFpypQp2W277LJLevLJJ9ssfzJ48OB04oknZsH5LbfcMsuGv/LKK7NlU84888z05ptv1vmVAN0hAA8AAAAA9HmyEAGolTPOOCOde+65aeHChWnFFVdMN998c7rxxhvTCius0O7jI+A+derU7Lg0fPjwNG/evCwwf/nll9e87UDPDarAcwBAw3riiSfScsstV+9m0I+98sor2WxnAACgflmIp512WrrggguyQEghC/Giiy7KAiE/+9nP2mQh7r777unggw9O99xzT5aFeMMNN6Tjjz8+HXnkkWnIkCF1fjUA9AbHHHNMuuyyy9Iee+yRJk2alJZZZpkut4ms+MiO32mnnbIlDB9++OHseAT0PgLwAPRpEXxvamqqdzMAAACocxZiiCzE73//+2nnnXfu8PGFLMQInBx77LFp9uzZWWA+Mua/8pWv1LDlAPRWI0eOTE899VS3xiVjclhky8cyKYsvvnhV2gdUlxL0AAAA/cTWW2+dZVUccMAB9W4KANQ0C3HEiBHZmu+x9m5nwffiLMSoqhbZ8quttposRADK0tOkIElF0HvJgAcAAAAA+ixZiAAA1JIMeAAAAACgT5OFCABArciABwAA6EMiQ68nQYKebg8AAEDnpk6dWu8mAFUkAx4AAKCPePXVV9PYsWPTYYcdlubOnVvWti+99FLabbfd0iabbJIWLlxYtTYCAAAA9GUC8AAAAH3EpEmTsiD8JZdcktZaa600ZcqULrdpaWlJl112Wfb4WON2+vTp2XUA6E9ZiHE8vPLKK+vdFAAA+gABeAAAgD7iuOOOSxMnTkyDBw9OL774YpowYUKW1T5z5sx2H//Pf/4zbbPNNumQQw5Js2fPTkOHDk2nnnpqOuigg2redgAAAIC+QAAeAACgjxg+fHg677zz0uOPP5522mmn7LbIah83blyW1R7ZfeHtt99OZ555ZlpvvfXS3XffnQYMGJD222+/9NRTT6WTTjopLbnkknV+JQAAAAC906B6NwAAAIDKWnPNNdMtt9yS7rjjjnTkkUemxx57LMtyj0B7uPbaa1uD8ZtttlkWtN94443r3GoAAACA3k8GPAAAQB+13XbbpYceeihdfPHFqampqTXoHv8fM2ZMFoi/9957Bd8BAAAAKkQAHgAAoA+LrPeBA9s/9evodgAAAAC6x2gLAABAH3XnnXemDTbYIB166KGpubm5tQR9/P/ZZ59Ne+65Z9piiy3Sgw8+WO+mAgAAAPQJAvAAAAB9zPTp09OECRPStttumx599NE0bNiwdOmll2bB9rD33nun008/PS2xxBKtJegPPPDANHPmzHo3HQAAAKBXE4AHAADoI+bOnZuOPvroNG7cuDRlypTstl122SU9+eST6ZBDDmnNgB88eHA68cQTs+D8lltuma0Jf+WVV6Y111wznXnmmenNN9+s8ysBAAAA6J0E4AEAAPqIM844I5177rlp4cKFacUVV0w333xzuvHGG9MKK6zQ7uMj4D516tQsO3748OFp3rx5WWD+8ssvr3nbAQAAAPoCAXgA+oyRI0emWbNmtbnEbVBPvpdALR1zzDFpxIgR2ZrvkfW+8847d7lNZMVHdvwTTzyRZcuvttpq6eCDD65JewEAAAD6mkH1bgAAVMrAgQNTU1NTvZsBbfheArUUE3yeeuqpbv3uRJZ8ZMs3NzenxRdfvCrtAwAAAOjrBOABAAD6kJ5O+jFpCAAA6Mg777yTXnvttXo3o8+ICmaDBgnVQV/jrxoAAAAAAIBOXX311enwww9Pc+bMqXdT+oxhw4aliy66KO277771bgpQQQLwAAAAANDLyECsLBmIAF0fdwTfKy/ez3hf99prL8ch6EP8NQMAAPQTU6dOrXcTAKgAGYiVJwMRoHMx6ctxpzrifY3313Jg0HcMrHcDAAAAAIDSyECsbgZivL8AANATMuABqqi5ubnq+zAzEgAAoP+QgVg9MhAByvPEE0+k5ZZbrt7N6HVeeeWVNG7cuHo3A6giAXiAKho1alTV99HS0lL1fQAAAAAA5EXw3aQlgEUJwAMAANAwtt5663T33Xen/fffP1155ZX1bg5AryADsXtkIAIAUA0C8AAAAADQi8lABACAxjGw3g0AAAAAAAAAgL5ABjxAFc2aNavT+2fMmJHGjx/f7n3Tpk1Lo0ePrlLLAADqp7m5uUeZmj3dHgAAAKBaBOABqqirgeH58+d3eN/IkSMNLAMAfc6rr76axo4dm/bYY4909tlnp2WWWabkbV966aV0+OGHp4cffjhb73jxxRevalsBAAAAyqUEPQAAADUzadKkLAh/ySWXpLXWWitNmTKly21aWlrSZZddlj3+xhtvTNOnT8+uAwAAADQaAXgAAABq5rjjjksTJ05MgwcPTi+++GKaMGFC2m233dLMmTPbffw///nPtM0226RDDjkkzZ49Ow0dOjSdeuqp6aCDDqp52wEAAAC6IgAPAABAzQwfPjydd9556fHHH0877bRTdltktY8bNy7Lao9s9/D222+nM888M6233nrp7rvvTgMGDEj77bdfeuqpp9JJJ52UllxyyTq/EgAAAIBFWQMeAACAmltzzTXTLbfcku6444505JFHpsceeyzLco9Ae7j22mtbg/GbbbZZFrTfeOON69xqAAAAgM7JgAcAAKButttuu/TQQw+liy++ODU1NbUG3eP/Y8aMyQLx9957r+A7AAAA0CsIwAMAAFBXkfU+cGD7p6cd3Q4AAADQiIxkAAAAUDd33nln2mCDDdKhhx6ampubW0vQx/+fffbZtOeee6YtttgiPfjgg/VuKgAAAECXrAEPANBNESiqtijHDNAXTZ8+PR111FFpypQp2fVhw4alSZMmpWuuuSbdc889ae+9907jxo1Lp512WmsJ+v333z+deeaZaYUVVqh38wEAAADaJQAPANBNo0aNqvo+CmshA/QVc+fOzYLqF1xwQVq4cGF22y677JIuuuiiLLD+s5/9LLtt8ODB6cQTT0y77757Ovjgg7Og/JVXXpluuOGGdPzxx6cjjzwyDRkypM6vBgAAAKAtJegBAAComTPOOCOde+65WfB9xRVXTDfffHO68cYbO8xqX3PNNdPUqVPTpZdemoYPH57mzZuXBeYvv/zymrcdAAAAoCsC8AAAANTMMccck0aMGJGt+f7kk0+mnXfeucttYj34Qw45JD3xxBNZtvxqq62WZcUDAAAANBol6AEAAKiZkSNHpqeeeio1NTWVvW1kyUe2fHNzc1p88cWr0j4AAKD9fvysWbMWuY3yeS+h7xOABwDopuKTpWIzZsxI48ePb/e+adOmpdGjR1epZQCNrTvB90puDwAAlGfgwIH64RXivYS+TwAeAKCbujpZmj9/fof3xcxmJ1sAAJRL1lzleC8ByvfOO++k1157rd7NgEXEUmeDBgl70hh8EwEAAACgl5A1VzneS4DyXH311enwww9Pc+bMqXdTYBHDhg1LF110Udp3333r3RQQgAcAAKBxTJ06td5NAOgVZCDSqGQgQt897gi+08jiuxnf0b322stxiLrzDQQAAACAXkQGIo1MBiL0TTHpy3GHRhff0fiuqnBDvQ2sdwMAAAAAgNLIQKS3ZCDGdxUAoD+SAQ8AAAAAvYQMRHoDGYjQPzzxxBNpueWWq3cz6MdeeeWVNG7cuHo3AxYhAA8AAAAAAEBZIvhuog3AogTgAQAAAKAXk4FIvclABAD4HwF4AAAAAOjFZCACAEDjEIBvINOnT08PPPBAev7559PChQvTiBEj0tixY9Omm26ahgwZUrd2tbS0pL/97W/p4YcfTrNmzcpuW3755dN6662XNtxwwzRgwICK7/Oll15Kf/3rX9PTTz+dXn/99TR48OC07LLLpjXWWCPbb7w3ve01AQAAAAAAAH2bAHwDmDJlSjrttNOygHB7ll566XTAAQekk08+uablxN5+++10/vnnp8mTJ6cXXnih3cestNJKaeLEielrX/taFiTviffeey9dc8016fvf/376y1/+0uHjIjgeJa123HHHdOKJJ6ZlllmmYV8TAAAAAAAA0H8MrHcD+rO33nor7bvvvmnChAkdBt/DvHnz0kUXXZQFne+5556atO25555LH/3oR9PRRx/dYaA6RLb+UUcdlTbZZJNOH9eVf/zjH+kjH/lI+sIXvtBp8L2Qvf73v/89TZo0Kb344osN+5oAAAAAAACA/kUAvk4i23vPPffMMr7zFltssbTqqqum9ddfPw0bNqzNfc3NzWmHHXZI999/f1XbFiXZt9lmm/TQQw+1uX3JJZdMa6+9dlprrbUWKYn/4IMPZtu88sorZe/v7rvvzgLj8RzF78XKK6+cNtpoo7TBBhtk/+4trwkAAAAAAADofwTg6+Scc85Jv/zlL9vc9n//939pxowZ6T//+U8WKH711VfTTTfdlEaPHt36mAULFqQ99tgjzZkzp2pti3L3sR59QQSmo2R7BKIff/zx9MQTT2T//t73vtcmaP2vf/0rHXTQQWXt65FHHkk77bRTmjt3buttEQz/6U9/mgXN4/2YNm1aViEg/h2v+ze/+U067LDD0vDhwxvyNQEAAAAAAAD9kwB8Hfz3v/9NZ5xxRpvbvvOd76SLL744rbjiiq23DRw4MCtPf99996VVVlmlTYn0CBRXw+23355++9vftl6PNdBvu+229PWvfz0ttdRSrbcPHTo0HXHEEel3v/tdm3XSf/WrX6W77rqrpH0tXLgwff7zn0+vv/56623xnBGUj9L8yy677CLbxHrvUQXgBz/4QVZ+Pv++NMJrAgAAAAAAAPovAfg6iLXL80HnLbfcMh177LEdPv4DH/hA+tGPftTmtvPOOy8L5Ffat771rTbXjzvuuKx9Hdlqq60Wafs3v/nNkvYVkw4i87zgy1/+cjaxIB/87kyUjy8uG1/v1wQAAAAAAAD0XwLwdVj7/Yorrmhz2ymnnJIGDBjQ6Xbbbrtt2mKLLVqvRwD/uuuuq2jbHnvssfTAAw+0yQg/+uiju9zumGOOyR5bEBn7Tz75ZKfbRHn5CMAXjBkzJivLX2m1fE0AAAAAAABA/zao3g3obyKQ29zc3Hp9tdVWS1tvvXVJ237xi19Mf/zjH1uvT5kyJVsLvVKK16SPtebf9773dbldPGb33XdPV155ZZu2xVruHbnqqqvSW2+91Xo9guL5cvC98TUBANB/vfPOO+m1116rdzP6jBEjRqRBg5yuAgAAAL2PEY0a+/Wvf93m+ic+8Ykus9/zj82bOnVqmj9/fptM7Uq2bfvtty9522hbPlh96623puOPP77Dx19++eWt/46BtT333DNVQy1fEwAA/dPVV1+dDj/88DRnzpx6N6XPGDZsWLrooovSvvvuW++mAAAAAJRFAL7GHn744TbXN91005K3XXHFFdMqq6ySnnnmmez6woULszXUN9544x63q6WlJT366KPdbttmm23W5vojjzySPWd7kwumT5+ennrqqdbr66yzTlpuueVSpdXyNdE7rHJc2wkZjeCdOS93eN/mZ9+ZBg1bPjWiZ87asd5NAICGyXwXfK+8eD/jfd1rr71kwgMAAAC9ijXga6x4HfFx48aVtX3x4yu1Lvmzzz6bFixY0Ho9supHjx5d8vaxhnu+hHxk5j/33HPtPvavf/1rm+vrrbde679feOGFdOaZZ6aPfexjaYUVVkhDhgxJH/jAB7LrJ5xwQvrb3/7WkK8JAID+KcrOC75XR7yvyvoDAAAAvY0AfA298cYbacaMGW1uW3nllct6juLH5zPJe6L4ecptVzltKw7Ar7baallm+fnnn59WX331dOKJJ6a//OUv6aWXXsrWiX/xxRez69/5znfSRhttlK3NHvc10msCAAAAAAAAUMuvhl555ZUs0FwwePDgNGrUqLKeI7LB82bNmlWRthU/z0orrVT2c0Tb8gHqjtr273//u831ZZZZJh166KHpsssuK2k/N9xwQxbE/93vfpfGjh3bEK+pXPE8zc3NZW1T/L4BANCYYpmoaiyx1B/Ol8qtEAYAAADQaATga2jevHltrkd583LXE48y6p09Z6XaVryfSrZt9uzZba7/5Cc/aVNafsMNN8zWevzgBz+YXf/nP/+Zfv7zn6eHHnqoTXn5T3/60+nhhx/OAvj1fk3l+sEPfpBOPfXUijwXAACNJYLvTU1N9W4GAABAxYwcOXKRBLW4DerJ95JGJQBfQ8XB21jfvFxLLrlkp8/ZG9pWHIAvBN8HDhyYLrjggvTlL395kYkJRx99dJo8eXL6xje+0VpF4Omnn04TJ05MP/7xj+v+mgAAAAAAoK+K8XsTjWk0vpc0KmvA19Cbb77Z5vriiy9e9nMsscQSi6wr39va1lEQ++yzz05f+cpX2q0KELcdccQR6dvf/nab23/605+mGTNm9Lr3GwAAAKAnmV75i0wv6s33EgDgf2TA11BxBvbChQvLfo633nqr0+fsDW1r7/a11lorHXnkkV3u47jjjktXXXVV63ro77zzTpYBf8opp/Sq9zuy/HffffeytonXvPPOO1dk/wAAAEDvJNOLRuR7CQDwPwLwNbT00kt3mqFdiuIM7OLn7A1ta+/2L37xi1lHvSuDBg1KBx10UDrhhBNab7v77rt73fs9atSo7AIAAAAAAAD0HUrQ11Bx8HbBggWt65mXav78+Z0+Z6XaVryfSratvdu32mqrkvdT/Nhp06bV/TUBAAAAAAAAyICvoeWWWy5by7wQdH/77bez9ZCWX375kp/jhRdeaHO9UlnUxc/z/PPPl/0cpbatvde75pprlryfD33oQ4usKR+Z6ksuuWTdXhMAAAAAALXT3Nxc9X1YWgGA7hCAr6EIEI8ePTo9++yzrbfNmDGjrAB8PD5v7NixFWlbcVD7ueeeK/s5irfpqG2x3nteTEp43/veV/J+lllmmUVue+211xYJwNfyNQFQf6ecckpqNLNnz+7wvsmTJ6fhw4enRtSI7yUAAADUOlmq3Aq2ABAE4GssArj5APwTTzyRNt5445K3f/LJJxd5vkoYM2ZMFsAurHkepdejnXF7KeKxUVK/YOjQoWnllVdu97Hjxo1bpBOzcOHCtMQSS5S0r/bWcl9qqaXq+poAAACgv5KBCAAA8D8C8DW2/vrrp9tuu631+n333Zf233//kradOXNmeuaZZ1qvDx48eJFgdndFFvq6666b/vKXv7RpW6nB6j/96U9trsdzxXO2Z8MNN1zktpdffjmrDlCKKNuft9hii6Vhw4bV9TUBAABAfyUDEQAA4H8G5v5NDXzmM59pc/2OO+4o+STy9ttvb3N9m222SUsvvXTV2vb73/++5G2LH7vTTjt1+NhVV101ffjDH25z24MPPljyvoofG+vHdxQYr9VrAgAAAAAAABCAr7FNN900Lbfccq3X//Of/6SpU6eWtO3ll1/e5vrnPve5irbts5/9bJvr119/fZo3b16X273++uvZY8tp24QJE9pcv/HGG0tuZ/G+tt5664Z4TQAAAAAA1EZUSu3sMm3atA63jfu62r64EisAlEoAvsYGDhyYDjjggDa3nXrqqV1mwf/hD39If/zjH1uvv+9970t77LFHRdsWJdbz69FHoHrSpEldbhePifXVCz72sY91WRr/wAMPzEroF0Sw+x//+EeX+3r00UfTlClT2tzW2ftQy9cEAAAAAEBtNDU1dXoZOXJkh9vGfV1tHxcA6A4B+Do49thj25SOv/vuu9PZZ5/d4eNfeOGF9KUvfanNbV//+tfbZNK3J8qy5y+lZNp/+9vfbnP9rLPOSvfcc0+Hj2+v7aeffnqX+4ky9Iccckjr9YULF6a99torvfrqqx1u09zcnPbee+/0zjvvtAmMd5YBX8vXBAAAAP2RDEQAAID/EYCvgwicn3DCCW1uO/7449OXv/zl9OKLL7be9t5772XZ3lG2/plnnmm9fcUVV0zf+MY3qtK2T33qU2n77bdvvf7222+nT37yk+n8889PCxYsaL09ssMnT56cPT4eU/DpT386bbvttiXt66STTmozi/CRRx5JH/3oR9Mtt9zSJsgezx/vQ9z3xBNPtN4+ZMiQdMkllzTUawIAAID+RgYiAADA/wzK/ZsaZ8Hfd9996dZbb2297eKLL04//OEP05gxY9KwYcPS008/nWbPnt1muyWXXDJdd911afjw4VVr209+8pO0ySabZPsPb775Zpo4cWI2SWC11VbLyuXH2vVxe97qq6+errzyypL3M2rUqOy17LDDDq3P9e9//ztbaz1e/yqrrJLdFu2YO3dum20XW2yx7L1ab731Guo1AQAAAAAAAP2XDPg6rgUf655H2fW8d999NwsEP/TQQ4sE32NW+G9+85u02WabVbVtyy+/fLrrrrsWCW6/8cYb6e9//3uWhV4cqF5//fWzbcqdlR7l42+77bYsGJ83Z86cLCM+LsXB92WWWSbLiN9vv/0a8jUBAAAAAAAA/ZMAfB1FCfVrr7023XDDDVmwtyNDhw7NytNHkLir9c4rJbLwH3jggWwt9Ch535G4b9KkSekvf/lLWnnllbu1ry233DI9+eST6aijjuo0s3/ZZZfNstYjS/4zn/lMQ78mAAD6h5gkW7xOcWellumY9xIAAADoC5SgbwC77rprdonAcgR9X3jhhbRw4cIsGL3WWmtlGe8RrC9XlFXvicUXXzwdc8wxWWD8wQcfzLLRYxAsRMZ6TBrYcMMNs2z+norg+jnnnJPOPPPMLEgeWemvvPJKGjx4cJaBPnbs2PSRj3ykx/uq5WsCAKDvi36jikmV4b0EAAAA+gIB+AayxhprZJdGHAjbeOONs0u1RcA9JhxUu8x+LV8TAAB93zvvvJNee+21ejcDFjFixIg0aJBTfwAAAKgVZ+EAAAA9cPXVV6fDDz88zZkzp95NgUUMGzYsXXTRRWnfffetd1MAAACgX1BnGwAAoAeZ74LvNLL4bsZ3NL6rAAAAQPUJwAMAAHRTlJ0XfKfRxXfUEgkAAABQGwLwAAAAAAAAAFAB1oAHAACooCeeeCItt9xy9W4G/dgrr7ySxo0bV+9mAAAAQL8kAA8AAFBBEXxvamqqdzMAAAAAqAMl6AEAAAAAAACgAgTgAQAAAAAAAKACBOABAAAAAAAAoAIE4AEAAAAAAACgAgZV4kkAaN+7C+Z0fv8br3d634DBQ7rcx2JLDetW2wAAAAAAAKgsAXiAKnr+wn26ve1LV00s6XFjjr212/sAAAAAAACgcpSgBwAAAAAAAIAKEIAHAAAAAAAAgAoQgAcAAAAAAACACrAGPEAVrfTVa+rdBAAAAAAAAGpEAB6gihZbali9mwAAAAAAAECNCMADAHTT/PnzO71/wYIFnd43ePDgLvcxdOjQbrUNAAAAAIDaE4AHAOimc889t9vbXnbZZSU97uSTT+72PgAAAAAAqK2BNd4fAAAAAAAAAPRJAvAAAAAAAAAAUAEC8AAAAAAAAABQAdaABwDopqOOOqreTQDqbOTIkWnWrFmL3Ab15HsJAAAA9SMADwDQTUOHDq13E4A6GzhwYGpqaqp3M6AN30sAAACoHwF4AAAAgD5ileN+nRrNO3Ne7vC+zc++Mw0atnxqRM+ctWO9mwAAAPRCAvAAAAAAAECfmfzVWyeAmfwF0DcMrHcDAAAAAAAAAKAvEIAHAAAAAAAAgApQgh4AAPqQ5ubmqu+jqamp6vsAAAAAgN5IAB4AAPqQUaNGVX0fLS0tVd8HAAAAAPRGStADAAAAAAAAQAXIgAcAAAAAAAAahiX26M0E4AEAAAAAAICGYYk9ejMBeAAA6ENmzZrV6f0zZsxI48ePb/e+adOmpdGjR1epZQAAAADQ9wnAAwBAH9JV+bT58+d3eN/IkSOVXwMAAACAHhjYk40BAAAAAAAAgP9PBjwAAAAAAADQMCyxR28mAA8AAAAAAAA0DEvs0ZsJwANAP9Pc3Fz1fejgAgAAAADQHwnAA0A/M2rUqKrvo6Wlper7AAAAAACARjOw3g0AAAAAAAAAgL5AAB4AAAAAAAAAKkAAHgAAAAAAAAAqwBrwANDPzJo1q9P7Z8yYkcaPH9/ufdOmTUujR4+uUssAAAAAAKB3E4AHgH6mqamp0/vnz5/f4X0jR47scnsAAAAAAOivlKAHAAAAAAAAgAoQgAcAAAAAAACAClCCHoBeobm5uer7UFodAAAAAADoCQF4AHqFUaNGVX0fLS0tVd8HAAAAAADQdylBDwAAAAAAAAAVIAAPAAAAAAAAABWgBD0AAAAAAEA/1dzcXPV9NDU1VX0fAI1CAB6AXmHWrFmd3j9jxow0fvz4du+bNm1aGj16dJVaBgAAAAC916hRo6q+j5aWlqrvA6BRCMAD0Ct0NUt2/vz5Hd43cuRIs2wBAAAAAICqE4AHAAAAoNveXTCn8/vfeL3T+wYMHtLlPhZbali32gYAAFBrAvAAAAAAdNvzF+7T7W1fumpiSY8bc+yt3d4HAABALQnAAwAAAAAA9FOzZs3q9P4ZM2ak8ePHt3vftGnT0ujRo6vUMoDeSQAeAAAAAACgn2pqaur0/vnz53d438iRI7vcHqC/EYAHAAAAAAB6lXcXzOn8/jde7/S+AYOHdLmPxZYa1q22AdC/CcADAAAA0G0rffWaejcBgH7o+Qv36fa2L101saTHjTn21m7vA4D+SwAeAAAAgG6THQgAAPA/A3P/BgAAAAAAAAC6SQY8AAAAADSA5ubmqu+jqamp6vsAAID+TAAeAAAAABrAqFGjqr6PlpaWqu8DoBZW+uo19W4CALRLAB4AAAAAAOhVFltqWL2bAADtsgY8AAAAAAAAAFSADHgAAKiSVY77dWo078x5ucP7Nj/7zjRo2PKpET1z1o71bgIAAAAAdEkAHgAAAAAawKxZszq9f8aMGWn8+PHt3jdt2rQ0evToKrUMAAAolQA8AABAO5qbm6u+j6ampqrvA4Deo6vjwvz58zu8b+TIkY4rAADQAATgASjfKcNSw5n9Xsf3TV4npeEDU0M6ZU69WwBAB0aNGlX1fbS0tFR9HwAAAADUToNGIwAAAAAAAACgdxGABwAAAAAAAIAKEIAHAAAAAAAAgAqwBjwAAEA7Zs2a1en9M2bMSOPHj2/3vmnTpqXRo0dXqWUAAAAANCoBeAAAgHY0NTV1ev/8+fM7vG/kyJFdbg8AANDGKcNSQ5r9Xsf3TV4npeENWGz5lDn1bgHQjzXgryIAAAAAAAAA9D4C8AAAAAAAAABQAQLwAAAAAAAAAFAB1oBvINOnT08PPPBAev7559PChQvTiBEj0tixY9Omm26ahgwZUrd2tbS0pL/97W/p4YcfTrNmzcpuW3755dN6662XNtxwwzRgwIC6tQ0AAAAAAACgUQjAN4ApU6ak0047LQtyt2fppZdOBxxwQDr55JPTcsstV7N2vf322+n8889PkydPTi+88EK7j1lppZXSxIkT09e+9rU0ePDgkp976tSpaZtttul228aMGZOeeeaZkh679dZbp7vvvrvb+7riiiuy9x8AAAAAAACgM0rQ19Fbb72V9t133zRhwoQOg+9h3rx56aKLLkrjxo1L99xzT03a9txzz6WPfvSj6eijj+4w+B4iW/+oo45Km2yySaePAwAAAAAAAOjrBODr5L333kt77rlnuuaaa9rcvthii6VVV101rb/++mnYsGFt7mtubk477LBDuv/++6vatigzH9npDz30UJvbl1xyybT22muntdZaa5GS+A8++GC2zSuvvFLVtgEAAAAAAAA0KiXo6+Scc85Jv/zlL9vc9n//93/pW9/6VlpxxRVbg/TxmCjxPmPGjOy2BQsWpD322CM9/vjjiwToKyXKrcd69AURbD/rrLPSwQcfnJZaaqnstvnz56cf/vCH6YQTTkhvvvlmdtu//vWvdNBBB6Vbbrml7H3ut99+6Qtf+ELJj4/JAN31+9//vqzHx6QDAAAAoJ85pTrjLj0y+72O75u8TkrDGzTX5pQ59W4BAADUjAB8Hfz3v/9NZ5xxRpvbvvOd76TjjjuuzW0DBw7MytN/5CMfSZtvvnnrmudR9v173/teOvXUUyvetttvvz399re/bb0e67rfdtttacstt2zzuKFDh6YjjjgibbjhhukTn/hEtl58+NWvfpXuuuuustd3X2211dJ2222XaqFW+wEAAAAAAOjNTjnllNSIZs+e3eF9kydPTsOHD0+NplHfSyqvQafF9m2TJk1Kr7/+euv1CG4fe+yxHT7+Ax/4QPrRj37U5rbzzjsvC+RXWmTg58WkgOLge95WW221SNu/+c1vVrxdAAAAAAAAAI1OAL7Goqz8FVdcsciMlwEDBnS63bbbbpu22GKL1usRwL/uuusq2rbHHnssPfDAA22y3I8++ugutzvmmGOyxxbcd9996cknn6xo2wAAAAAAAAAanQB8jUVwurm5uU3p9a233rqkbb/4xS+2uT5lypSKtq14TfpYa/5973tfl9vFY3bfffeqtg2gef57nV7+u6Clw23jvq62jwsAAAAAAEBPWAO+xn7961+3uR7rp3eV/Z5/bN7UqVPT/Pnz22SfV7Jt22+/fcnbRtuuvPLK1uu33nprOv744yvSLoAw6tx53d52/GXzS3pcy8nLdHsfAAAAAAAAAvA19vDDD7e5vummm5a87YorrphWWWWV9Mwzz2TXFy5cmJ544om08cYb97hdLS0t6dFHH+122zbbbLM21x955JHsOUudXAAAAJ06ZVhqOLM7qZ4yeZ2UhjdowbFT5tS7BQAAAAB9lgB8jRWvjT5u3Liyto/HFwLwheerRAD+2WefTQsWLGi9Hln1o0ePLnn7MWPGpKWWWqr1OSIz/7nnnivrOSJg//TTT6dZs2ald999Ny277LLp/e9/fxoxYkSqtDlz5mSvefbs2WnppZdOI0eOTCuttFJabLHFKr4vAAAAAAAAoH8QgK+hN954I82YMaPNbSuvvHJZz1H8+KeeeqoibSt+nnLbVdgm/zzx71ID8FdddVW68MIL06uvvrrIfWPHjk0f//jH05e//OW09tprp57aYIMNsmz/995rm7EUgfjI5N91113TF77whbTEEkv0eF8AAAAAAABA/yEAX0OvvPJKluVdMHjw4DRq1KiynuMDH/hAm+uRLV4Jxc8T2eDlirblA/DltC2f1V/sH//4R3a5+OKLs+D4pZdemmXHV2oZgIJ58+al2267LbucdNJJ6YILLki77757qoZ4b5qbm8va5t///ndV2gK9xayjlq53EwAAAACgz2me38nyWiml/y5o6fS+oYM73z40DW3QJboAqkAAvoYiwJsXJdvLXSM9SsN39pyValvxfurZtoKYvHDDDTekBx54IP32t78tu3x/OV566aW0xx57pKOOOiqdc845FX/+H/zgB+nUU0+t+PNCX6aTDgAAAACVN+rc7o/lj79sfkmPazl5mW7vA6C3Ec2ooeKA9JAhQ8p+jiWXXLLT5+xtbVtzzTXTkUcemX75y19mGd5z585Nb7/9dpYhfu+996aTTz45Wwc+L8r4f/rTn04vv/xyyW2L17PTTjtlge/77rsve/6FCxem119/PU2fPj1dffXVaccdd1xkQsS5556bzjrrrJL3AwAAAAAAAPRfMuBr6M0332xzffHFFy/7OYrXJY915Xtj21ZZZZV01113pa233rrd+5uamrJLrMl+3HHHpYkTJ2al5wueffbZbE34G2+8sct2RYA/nmfkyJGL3BfLAMTa76uttlraZ599sqD/XnvtlV544YXWx5xwwglphx12SOutt16X+wIAAAAAAAD6LwH4GirOKo8M7HK99dZbnT5nb2lbBODjUmrbLrnkkizDfvLkya2333TTTWnatGlp/PjxnW7/2c9+NpVq8803T1OnTk2bbLJJeuWVV1pL33/zm99Mv/rVr1KlxOSBcteXjwoBO++8c8XaAAAAANBXNTc3V30fkTwCAADFBOBrKDKtO8s6L0VxVnnxc/bFthXEWuy33nprFoguiNLxXQXgy7XGGmtk+zrwwANbb/vNb36TXn311bTssstWZB+jRo3KLgAAAABUXi3GXSJpA6AvmHVUZcfyAfo7AfgaKg5IL1iwIOuoF6873pn58+d3+pyValvxfurZtoJBgwalr33ta9ml4Pbbb0/V8IUvfCEdc8wxrbOl33vvvXTHHXekPfbYoyr7AwColHcXzOn8/jde7/S+AYO7rrC02FLDutU2AKBzzfPf6/T+/y5o6fS+oYM73z40DR3YrbYB0Hc5NgBUlgB8DS233HJZsL0wO/btt99Os2bNSssvv3zJz5Ffm7ySs3mLn+f5558v+zmq1ba8bbfdts31f/3rX2VPYijFwIEDs/Xpr7/++tbbnnrqqYruAwCgGp6/cJ9ub/vSVRNLetyYY2/t9j4AgI6NOndet7cdf1lpyRQtJy/T7X0AAABdM62phmIN89GjR7e5bcaMGWU9R/Hjx44dW5G2fehDH2pz/bnnniv7OYq3qVTb8lZeeeU2199555302muvVXw/7e2rFmuHAQAAAAAAAL1Xv82Aj/Lv+bXE11133ZrsN4LSzz77bOv1J554Im288cYlb//kk08u8nyVMGbMmGyCQGEd9ygnH+2M20sRj433tGDo0KGLBLArYfDgwYvcFpUEqqF4X9XaDwAAAACVFVUnu0pyGT9+fLv3TZs2bZEkGgAA6JUB+A033LD135deemmXgemeBNHvv//+tP3222f/jvLlkUldC+uvv3667bbbWq/fd999af/99y9p25kzZ6ZnnnmmTYB43LhxFWlXvAfx/v3lL39p07ZSA/B/+tOf2lyP56p0Wfjw0ksvtbke+xg5cmTF99PevpqamqqyHwAAAAAqq6txnEg+6UiMNRkHAgCgTwTgH3744dag6uuvv171IHphLfZa+sxnPpPOPvvs1ut33HFHyWuY33777W2ub7PNNmnppZeuaNvyAfjf//73ae+99y5p23hs3k477ZSq4d57721zfYUVVkiDBg2qyb6qkdEPAFBpK331mno3AQDopllHVW6cBwAAqI+GCsCHcrOm6xFE74lNN900LbfccumVV17Jrv/nP/9JU6dOzYLpXbn88svbXP/c5z5X0bZ99rOfTd/61rdar19//fXpggsu6DLIH5Ml4rHVbFtH78G2225blf3cfffdafr06TXZFwBAJS221LB6N6HPaJ7/Xqf3/3dBS6f3DR3c+fahaejAbrUNgL7JcQEAAHq/hgvA93UDBw5MBxxwQDr33HNbbzv11FPT1ltv3enkgz/84Q/pj3/8Y+v1973vfWmPPfaoaNuibHyU/f/rX/+aXZ83b16aNGlS+va3v93pdvGYfNmuj33sYxUrjZ939dVXZ5MV8nbeeeeK7ydey9e+9rU2t62zzjpptdVWq/i+AABoXKPOndftbcdf1nFZ27yWk5fp9j4AAAAAaDym1dbBscce2yarPLKt82Xpi73wwgvpS1/6Upvbvv71r2eZ9J2JgH7+Uhy8bk9xsP2ss85K99xzT4ePb6/tp59+eqf7+PnPf55uuummsqoXXHvttYu8B+uvv36aMGFCp9vF+/Tiiy+WvJ+oTBCVAB599NE2t8ckCQAAgHprbm6u+gUAAADoPgH4OojA+QknnNDmtuOPPz59+ctfbhMsfu+999KUKVOysvXPPPNM6+0rrrhi+sY3vlGVtn3qU59K22+/fev1t99+O33yk59M559/flqwYEGbLPHJkydnj4/HFHz605/uslT7P/7xj7TrrrumNddcM5144onp/vvvb5NBX7Bw4cJ05513ZuXsP//5z6e33nqr9b4hQ4akiy++uMslC6KEfmSuR6D+mmuuafM+5j333HPpnHPOyTLdY5/FWfZdBfoBAABqYdSoUVW/AAAAAN2nBH0ds+Dvu+++dOutt7beFgHlH/7wh2nMmDFp2LBh6emnn06zZ89us92SSy6ZrrvuujR8+PCqte0nP/lJ2mSTTbL9hzfffDNNnDgxmyQQwezIXI+16+P2vNVXXz1deeWVJe/n3//+dzrzzDOzS5TmX2mllbLXFa9xzpw5WbC8eB9h8ODBWTA9St2XIgL3MZEhLmGZZZZJK6ywQvYex+SBl19+ucMs+S222CL97Gc/K/k1AQAAAAAAAP2XAHydRMD5+uuvTwceeGBWkr3g3XffzYLb7Rk5cmS64YYb0mabbVbVti2//PLprrvuyjLPH3nkkdbb33jjjfT3v/+93W2iHPwtt9ySmpqaurXPyPafMWNGdulMZM1HQHyjjTZK3TV37tzs0tXnc9RRR2Xl9CPgD1AtXVXyaDSrrrpqalQnn3xyvZsA9DGzjvrfslEAAAAAUAoB+DqKMuqxtvluu+2WBXoffvjhdh83dOjQtP/++2eBhVqVA4ws/AceeCArMx/l5zvKEI9y+JEdH2utL7744iU99x577JGVl4816eM1R2C/M4MGDcoy8g877LDsvSonIB4VBaKk/J/+9KeszHxX3v/+96c999wzHX744WmNNdYoeT8AAPQ9TUOt2AUAAABAeQTgG0Cshx6XKMn+l7/8Jb3wwgtZgDrKsa+11lpZxnsE68sVpeJ7IgLqxxxzTJYJ/uCDD2bZ8LNmzcrui4kAkfW+4YYbZtni5Rg3blxWdr6Q8f/UU09lWf/PP/98lpker33ppZdOI0aMyDItN95446wsfXccfPDB2SX897//TU8++WR69tlnU3Nzc7bu/GKLLZbtZ7nllksbbLBBVmIfAACgURXOyToSVcXGjx/f7n3Tpk1Lo0ePrlLLAAAAgCAA30Ai47oRs64jwB5B8LhUWgTAIyAfl2qLEv6bb755dgEAAOiNulr2KyYad3ZO1N1lwwAAAIDSqKkIAAAAAAAAABUgAA8AAAAAAAAAFSAADwAAAAAAAAAVIAAPAAAAAAAAABUwKDWob3zjG2nEiBGdPua1115rc/3jH/94yc9fvC0AAAAAAAAA9LkAfEtLS3r00UfL3ubuu+8ua5sBAwZk2wEAAAAAAABAnwzAR2C8Go8FAAAAAAAAgH4TgJeRDgAAAAAAAEBv1FAB+CuuuKLeTQAAAAAAAACA3h+A33///evdBAAAAAAAAADoloHd2wwAAAAAAAAAyBOABwAAAAAAAIAKEIAHAAAAAAAAgAoQgAcAAAAAAACACujzAfjXX389vfzyy+mdd96pd1MAAAAAAAAA6MMGpT6mpaUl/fznP09XX311uvfee9O8efNa73v/+9+fPvnJT6YvfOELaeutt65rOwEAAAAAAOg7mpubq76Ppqamqu8D6EMB+Aie33rrrdn/w2KLLZZ23HHHkrd/+umn06677poeeeSR1ufLmzlzZrrqqquySzzuxz/+cVp66aUr/CoAAAAAAADob0aNGlX1fRTHvvqq+fPnd3r/ggULOr1v8ODBXe5j6NCh3Wob9KoA/F//+tf0uc99Lg0YMCC7Hv8uNQA/Y8aMtOmmm6ZZs2a1/vgUnievcN+NN96YXnjhhXT77bf7AwMAAAAAAIAGce6553Z728suu6ykx5188snd3gf0mjXgf/vb37YJkh9xxBElbRePj4z2WOu9EHiPS9xefMnf9+c//zkde+yxVXxFAAAAAAAAAPQXDZUBf/fdd7f+e/XVV09bbLFFSdtdeeWV6cEHH2yT8T5w4MC0//77pz322COtssoqWamKyLC/8MIL09///vfWIPyll16aDjvssLT22mtX5TUBAAD0d9ZBBOjf2qtS2chWXXXV1Kj6S9lhAIDebFAjdR7zQfTddtut5G2/973vtXmeQYMGpZtvvnmR8vUbbLBBOvDAA9M+++yTbrjhhuy29957L1sL/rvf/W7FXgsAAAD/Yx1EAACgP4hlkrtaTnn8+PHt3jdt2rQ0evToKrUM6JcB+P/85z/p9ddfz/4dQfgddtihpO0iaJ/PaI//H3nkkR2uHT948OD005/+NPshe/bZZ7NtfvGLXwjAAwAAAAAAULXKXFGtuSMjR45U2SvnqKOOqncToPcH4KdPn96mfPxGG21U1rrxBYsvvniX67ovscQS2R/u4Ycfnl2fOXNmevHFF9OKK67YrbYDAADQ/5xyyimp0cyePbvD+yZPnpyGDx+eGlEjvpcAAED9DB06tN5NgG4bmBpElN0o+MAHPpCWWmqpkrabOnVq678j+/2Tn/xkGjFiRJfbTZgwoXWb8Mgjj3Sj1QAAAAAAAADQYBnwc+fObQ2IL7vssiVtE+XjH3jggTbl57fbbruStl1hhRWydQhjPY7YLjLgAQAAqDzrIAIAAAD9RcME4N944402JeJL8eSTT6Z58+a1ZrGHzTbbrOR9RhC+MBBUWH8eAACAyrIOIgAAANBfDGzEtRzmzJlT0jaR/V68/vs666xT8j6HDBnS+u8FCxaUvB0AAAAAAAAANGwAvrBue5SSf/rpp9N7773X5Tb3339/m+vrrrtuGjSo9KT+2bNnt/57ySWXLKu9AAAAAAAAANCQAfgPfvCDrf9euHBh+vOf/9zlNrfffnub9d+32GKLsvb5yiuvtP572LBhZbYYAAAAAAAAABowAL/hhhtm2euF9dx/+MMfdvr4++67Lz377LNtbttmm21K3t9LL72U/vvf/7ZeHz16dNltBgAAAAAAAICGC8DHeuzbbbddls0el6uvvjrddtttHT7+W9/6VpvrkcG+/fbbl7y/e++9t831NddcsxutBgAAAAAAAIAGC8CHQw89NPt/ZMHHGvC77LJLOuecc9Jrr73W+pj//Oc/affdd0933XVXm/Lz++yzTxo8eHDJ+4ry9fn152XAAwAAAAAAANBnAvCf+9znsjLyhaD6G2+8kY477rg0atSotMIKK2T/j7Xib7rppkWy548//viS9/Pmm2+m6667LttHXDbffPMqvBoAAAAAAAAA+pOGCsCHK664Iq244orZvwsZ7u+++256+eWX0yuvvNJaor6wVnz8/+yzz27dphQ33nhjmjt3bvY85a4dDwAAAAAAAAC9IgAfpeDvueeeNHbs2NZAe3uXQiA+1oI//PDDS37+2OY73/lOawC/kHkPAAAAAAAAAH0qAB9WW2219Mgjj6Tzzz8/bbjhhq3B9sJliSWWSDvuuGP64x//mE455ZSynvvnP/95euKJJ1qz39dbb720yiqrVOmVAAAAAAAAANBfDEoNatCgQemrX/1qdnnttdfSCy+8kF5//fU0fPjwLEAfQfju2H777dPTTz/den3ppZeuYKsBAAAAAAAA6K8aNgCfN2LEiOxSCSNHjswuAAAAAAAAANDnS9ADAAAAAAAAQG/TKzLgAQAAAACovebm5qrvo6mpqer7AACoFQF4AAAAAADaNWrUqKrvo6Wlper7APqXAQMGpN5m1VVXTY3o5JNPrncToNdRgh4AAAAAAAAAKkAAHgAAAAAAAAD6Wgn61VZbrW6lSKZPn16XfQMAAAAAAADQNzRUAP6ZZ57JguG1XvOnN64FAgAAAABQbbNmzer0/hkzZqTx48e3e9+0adPS6NGjq9QyAIDG1FAB+HoExGsd7AcAAAAA6C2ampo6vX/+/Pkd3jdy5MgutwcA6GsaMgAfBMYBAACg9CBHWLBgQaf3DR48uMt9DB06tFttAwAAABo0AB/B98UWWyx9/OMfT1/4whfSDjvskF0HAACA/uzcc8/t9raXXXZZSY87+eSTu70PAAAA6O8atgT9e++9l+64447sMmrUqLTPPvtkwfh111233s0DAAAAAAAAgEUMTA3kxz/+cdpqq61as+ALl5dffjmdd955aYMNNkjrr79+9u+4DQAAgPInPDfaZdVVV+2wvXFfvdvX0QUAAACgoQPwBxxwQLrzzjvT008/nU499dS0xhprtLk/gvGPPfZYOuqoo9LKK6+cPvOZz6Trr78+LVy4sG5tBgAAAAAAAICGLUE/evTo9K1vfSu73HfffenKK6/MAu1z5szJgvCRafDOO++k3/72t9ll2LBhac8990z77bdf2nTTTevdfAAAAKiKmJAOAAAANK6GDMDnRUA9LhdeeGG6+eab009+8pP0+9//Pr377rtZMD7Mnj07/fCHP8wuq6++erZWfATjx4wZU+/mAwAAQMUMHTq03k0AAAAAeksJ+s4sscQSaa+99kq/+c1v0nPPPZfOPvvstPbaa7d5TATk//3vf6eTTz45C8Rvs802Wfb8vHnz6tZuAAAAAAAAAPqHXhOAz3v/+9+fjj766Gw9+GnTpqXDDz88jRw5ss1j3nvvvXTPPfekL37xi9njIyM+MucBAAAAAAAAoBp6ZQA+b8MNN0wXXHBBevHFF7MS9TvvvHMaNGhQa0Z8XBYsWJCuueaatMMOO2T/BgAAAAAAAIBK6/UB+IIIun/uc59LN910UxaMj6D8RhttlN03YMCA7P+FNeMBAAAAAAAAoNL6TAA+b/HFF09LLrlkdgEAAAAAAACAWvj/tdr7gMhuv+2229JVV12VbrnllvTmm2+2yX4HAAAA4P+1dydQVpR3/rhflkYQEFEEw44aZUkUcchEcIFBjTquScRljKLEM0pmkEQFoxjAZUTIGDQ5EomOjnGLoENcR9GAGw78ggEXGCOytxrAjZ1G4H/e+p++07eBphuq+96+/Tzn3EO/1VV1v3U9x7frfup9XwAAAKpTrQ/g33nnnfDQQw+FRx99NPztb3/LhPExeC+dcj6OhI/T01966aVh3333zXHFAAAAAAAAABSiWhnAr1y5MjzyyCNJ8B4D+F2t73788ceHSy65JAwcODA0b948B5UCAAAAAKRj9OjRId98+eWXu/zdhAkTwv777x/yUT5+lgBAYag1AXxJSUmYOnVqMsX8tGnTwtatW7NGu5c65JBDwo9+9KMkeO/SpUsOKwYAAAAAAACgLsn7AP6NN95IRrpPmTIlfPXVVzudYj6Obo+j3GPoftxxx+W4YgAAAAAAAADqorwM4BcvXpyE7r///e+Tn3c2xXz9+vXDySefnITu55xzTmjcuHGOqgUAAAAAAACAPAvgf/e73yXB+8yZM7NC97JTzH/rW99KQveLL744HHzwwTmrFQAAAAAAAADyNoD/53/+56yp5Uu1atUqXHTRRUnwfvTRR+esPgAAAAAAAACoFQF8WQ0aNAj9+/dPQvfTTjstFBUVJdvXrFlTLe+33377Vct5AQAAAAAAAKgb8jaA37ZtW3jllVeSV3WLo+6//vrran8fAAAAAAAAAApXXgbwpWu+l5+KHgAAAAAAAADyVV4G8OWD+Ook5AcAAAAAAACg4AL4jh071kjoDgAAAAAAAAAFHcAvWbIk1yUAAAAAAAAAwB6pv2eHAQAAAAAAAABlCeABAAAAAAAAIAUCeAAAAAAAAAAotDXgc2HlypVh/PjxyQsAAAAAgP+zfv36Cn+/YcOGCn9XVFS02/do2rTpHtUGAJCP6mwA/+mnn4Y77rgj/O53vwsbN24UwAMAAAAAlPPLX/5yj4+N371WxqhRo/b4PQAA8k2dC+CLi4vD2LFjw/333x82b94ctm/fHurVq5frsgAAAAAAAACo5epMAL98+fLwb//2b+HBBx8MJSUlSfAOAAAAAAAAAHUmgJ85c2Z45ZVXwsKFC8Pq1auT0ept2rQJxxxzTPj+978fDj744AqPX7ZsWbjlllvC73//+7Bly5ZM8F466j22DzvssJAPPvroozB79uywYsWK5CGBli1bhq5du4Y+ffqExo0b56yu+Bm9/fbbYe7cuWHlypXJtvjf4Kijjgq9evWqlTMIFOI1AQAAAAAAALmVtwH8tGnTwjXXXBPef//9nf4+jmQfNmxYuOqqq8K4cePCPvvsk/X7uK776NGjw91335014r1s8H7EEUeEG2+8MVx00UUhl6ZOnZo8JBAD4Z1p1qxZGDRoULIWUqtWrWqsrvjAwl133RUmTJiQTN2/M+3bt0/+OwwdOjQUFRVV+twzZswI/fv33+PaOnXqFJYsWZJX1wQAAAAAhebaa6/NdQkAALVKXgbwEydOTMLPbdu2ZU0VXzY8j77++uvwm9/8JgmuX3755UwIP2fOnHD++eeHxYsXZwXv8ef46tGjRxg5cmQYOHBgTkc6xzXoBw8eHB555JEK91u3bl1ynX/4wx/ClClTwgknnFAjU/afffbZ4S9/+UuF+8XR+vGP8Mceeyz88Y9/DO3atQv5qhCvCQAAAACqU9OmTXNdAgBArVI/5Jk333wz/Ou//mvYunVrEpbHgLxs8F42UC8N1eM09cOHD0+2x+nqTzzxxEz4XvbYOL345MmTw7vvvpsE9LkM3+PDBbGG8uF7gwYNQpcuXULPnj1DixYtsn63atWqcNppp4W33nqrWmuLU7LH0enlg+omTZokDy9069Zthynx40MP8Zi4TEA+KsRrAgAAAAAAAPJL3o2Aj1PKx3C6bHDeqFGj8Hd/93fJ1OCxHUcox3A0Ti1fGsL/9re/DRdffHH44Q9/GDZs2JAV0B955JHh5ptvDmeddVbIF+PHj09GV5d15ZVXhptuuim0bds2acfPIe4Tp0OPa9lH8driyP333ntvh4A+LXG6+7gefakYTI8dOzZcccUVYd999022rV+/PkyaNCnccMMNYdOmTcm2Dz/8MFx++eXh6aefrvJ7/uhHPwqXXHJJpfePwXm+XxMAAAAAAABQt+RVAB9HdsdguTQ4j/9ed911SSBaPmz+8ssvw2233RbuvPPOZL84Hf2ZZ54Zvvrqq8zxBxxwQBJ0x/A1l6Pdy/vss8+S2su6/fbbw/XXX5+1rX79+uHcc88N3/nOd8Jxxx2XWfM8PoAQr3vMmDGp1/bSSy+FF154IdOOa6C/+OKLO0x7H6ee+ulPfxp69eoVTj755GRt9eiZZ54J06dPr/L67occckg46aSTQnXI1TUBAAAAAAAAdUteTUEfg86oNHyP4fkdd9yx05He+++/f/L7cePGZaalj9OMlwbtcbr5uXPnhssuuyyvwvco1rx27dpMOwbBI0aM2OX+cQ3y++67L2vbr371qyTIT1scgV9WfCigojXn43T/5WsfOXJkyCeFeE0AAAAAAABA/smrAD5OK1/q8MMPDz/72c92e8w111wTjjjiiOTn0pHvrVu3DtOmTUumrM83cVr5Bx54IGvb6NGjd/uQwIABA8Lxxx+faccA/4knnki1tnfffTfMnj07a0R4nIFgd4YPH57sW2rmzJlhwYIFIR8U4jUBAAAAAAAA+SmvAvi//vWvyb8xjI5ruVdW3DcG76Uj5+M04q1atQr5KAa5q1atypp6vV+/fpU6dvDgwVntqVOnplpb+TXp41rzzZs33+1xcZ/zzjuvWmvbU4V4TQAAAAAAAEB+yqsAPq7fXurb3/52pY8rv+/pp58e8tVzzz2X1Y5rjVd2ivy4b1kzZswI69evr7baTjnllEofW762Z599NuSDQrwmAAAAAAAAID/lVQC/Zs2aTBjdsmXLSh8X14MvK44qz1dxXfqy+vTpU+lj27ZtGzp37pxpl5SUhPnz56dSV5w94J133tnj2vr27ZvVnjdvXnLOXCrEawIAAAAAAADyV/18Wx+9VIMGDSp9XPl9991335Cvyq8j3r179yodX37/tNYlX7p0adiwYUOmHdc/79ixY6WP79SpU9bnHkfmL1++vEo1xHB70aJF4X/+53/Cm2++mVzbF198EWrzNQEAAAAAAAB1R14F8IVu48aNYdmyZVnbOnToUKVzlN//gw8+SKW28uepal17W9t//ud/hlatWoVDDz00HHvsseG4445LHjY44IADQrdu3cJPfvKT8P7779eqawIAAAAAAADqloa5LqAuWb16ddYU5kVFRaF169ZVOke7du2y2itXrkyltvLnad++fZXPEWsrG1BXpbYlS5bs8nf/+7//m7wmTpwYfvCDH4R77703Cebz/Zp2V9uqVauqdMzChQtTeW8AAAAAAACgegjga9C6deuy2nF689I17ysrTqNe0TnTqq38++SytlLx4YUpU6aE2bNnhxdeeGG30/fn8zXdc889YcyYMamcCwAAAAAAAMgPeRvAz5s3LzRs2LDS+5b1+uuvZ400r4wTTjghVLfy4W3jxo2rfI4mTZpUeM7aVtvhhx8ezjjjjHDiiSeGHj16JDMCxPPEtd//+te/hmnTpiUj3j/99NPMMXEa/9NPPz3MmjUrtGnTJu+uCQAAAAAAAKib8jKAj+H5tddeu8fH9uvXr0rHxFHoX3/9dahumzZtymo3atSoyufYZ599dlhXvjbW1rlz5zB9+vRd/rc66KCDklffvn3D9ddfH4YNG5YE8aWWLl0ahgwZEp588sla+XkDAAAAAAAAhScvA/gYiFd1BHvZqdyremxNKT8Cu6SkpMrn2Lx5c4XnrC21xQA+vipb229/+9tkNPqECRMy25966qnw5z//Ofzd3/3dLo/L1887Pjxw3nnnVXkN+HPOOSeV9wcAAAAAAADqSAAfVXVt9D09tibD+mbNmlU4Qrsyyo/ALn/OQqyt1Pjx48Ozzz6bBNGlHn744V0G8Pl8TXGq/fgCAAAAAAAACkdeBfAdO3bcq+A935UPbzds2JA8AFCVa16/fn2F50yrtvLvk8vaSjVs2DAMHTo0eZV66aWXavU1AQAAAAAAAIUjrwL4JUuWhELWqlWrrOn1t2zZElauXBnatGlT6XMUFxdntdMaRV3+PCtWrKjyOaqrtrIGDBiQ1f7www93+RBDbbkmAAAAAAAAoDDUz3UBdUlcwzyO8i9r2bJlVTpH+f27du2aSm1HHHFEVnv58uVVPkf5Y9KqrawOHTpktb/++uvwxRdf1OprAgAAAAAAAAqDAL6GlQ9w58+fX6XjFyxYUOH59lSnTp2SBwTKTr2+dOnSSh8f941T6pdq2rTpDmF5GoqKinbYFmcSqM3XBAAAAAAAABQGAXwN69mzZ1Z75syZlT72k08+yZqmP4bR3bt3T6WuOIX7kUceuce1vfnmm1nteK6qrG1fWZ9++mlWO77HgQceWKuvCQAAAAAAACgMAvgadsYZZ2S1X3755cya8Lvz0ksvZbX79+8fmjVrVm21TZs2rdLHlt/3zDPPDNXhjTfeyGp/4xvfCA0bNqzV1wQAAAAAAAAUBgF8DevTp09o1apVpr1o0aIwY8aMSh17//33Z7XPPvvsVGs766yzstqTJ08O69at2+1xa9euTfatztp29RkMGDCg1l8TAAAAAAAAUBgE8DWsfv36YdCgQVnbxowZs9tR8K+88kp4/fXXM+3mzZuHgQMHplpbnGK9d+/emXYMqseNG7fb4+I+cX31Ut/97ndTmxq/rIcffniHhxXOOeecWn1NAAAAAAAAQOEQwOfAiBEjsqaOf/XVV8Mdd9yxy/2Li4vDj3/846xtV199ddZI+p2J65WXfVVmpP3NN9+c1R47dmx47bXXdrn/zmq/9dZbK3yPxx9/PDz11FOVnno/euyxx3b4DHr27BnOPffcvLgmAAAAAAAAAAF8DsTg/IYbbsja9vOf/zwMGTIkfPzxx5lt27ZtC1OnTk2mrV+yZElme9u2bcM111xTLbWdeuqp4ZRTTsm0t2zZEr73ve+Fu+66K2zYsCGzPY4OnzBhQrJ/3KfU6aefvttp4f/3f/83/OAHPwiHH354uPHGG8Nbb72VNdq8VElJSfjTn/6UTP1+0UUXhc2bN2d+17hx4zBx4sTkwYJ8uCYAAAAAAACAhrkuoC6Pgp85c2Z49tlnM9tioDxp0qTQqVOn0KJFi7B48eLw5ZdfZh3XpEmT8MQTT4T999+/2mp76KGHwrHHHpu8f7Rp06YwbNiw5CGBQw45JBm5Hteuj9vLOvTQQ8ODDz5Y6fdZuHBh+Ld/+7fkFafmb9++fXJd8Rq/+uqr5KGD8u8RFRUVhUceeSSZFj7frgkAAAAAAACou4yAz5EYOE+ePDlccMEFWdu3bt2aBMF/+ctfdgjfDzzwwPD888+Hvn37Vmttbdq0CdOnTw9HHXVU1vaNGzeG999/P8yfP3+HoDpOBx+POeigg/boPeNo/2XLloV33nknzJo1Kxklv7PwPY6ajyPmv//97+f9NQEAAAAAAAB1iwA+h+I06nFt8ylTpiRh7640bdo0mZ4+hsT9+vWrkdriKPzZs2cna6HHKe93Jf5u3LhxSWjeoUOHSp174MCBycjzOCI9jnbfnYYNG4bjjz8+PProo+G9994LxxxzTMi3awIAAAAAAAAwBX0eiOuhx1eckj2GvsXFxcn653E69m7duiUj3mNYX1VxWvW90ahRozB8+PBw7bXXhjlz5oR58+aFlStXJr9r3bp18tBAr169ktH8VdG9e/dk2vnSEf8ffPBBMup/xYoVYc2aNcm1N2vWLLRs2TJ06dIl9O7du1JBfS6vCQAAAAAAAEAAn0cOO+yw5JVvYhgdQ/D4SluDBg2SQD6+CuWaAAAAAAAAgLrJMF8AAAAAAAAASIEAHgAAAAAAAABSIIAHAAAAAAAAgBQI4AEAAAAAAAAgBQJ4AAAAAAAAAEiBAB4AAAAAAAAAUiCABwAAAAAAAIAUCOABAAAAAAAAIAUCeAAAAAAAAABIgQAeAAAAAAAAAFIggAcAAAAAAACAFAjgAQAAAAAAACAFAngAAAAAAAAASIEAHgAAAAAAAABSIIAHAAAAAAAAgBQI4AEAAAAAAAAgBQJ4AAAAAAAAAEiBAB4AAAAAAAAAUiCABwAAAAAAAIAUCOABAAAAAAAAIAUCeAAAAAAAAABIgQAeAAAAAAAAAFIggAcAAAAAAACAFAjgAQAAAAAAACAFAngAAAAAAAAASIEAHgAAAAAAAABSIIAHAAAAAAAAgBQI4AEAAAAAAAAgBQJ4AAAAAAAAAEiBAB4AAAAAAAAAUiCABwAAAAAAAIAUCOABAAAAAAAAIAUCeAAAAAAAAABIgQAeAAAAAAAAAFIggAcAAAAAAACAFAjgAQAAAAAAACAFAngAAAAAAAAASIEAHgAAAAAAAABSIIAHAAAAAAAAgBQI4AEAAAAAAAAgBQJ4AAAAAAAAAEiBAB4AAAAAAAAAUiCABwAAAAAAAIAUCOABAAAAAAAAIAUCeAAAAAAAAABIgQAeAAAAAAAAAFIggAcAAAAAAACAFAjgAQAAAAAAACAFAngAAAAAAAAASIEAHgAAAAAAAABSIIAHAAAAAAAAgBQI4AEAAAAAAAAgBQJ4AAAAAAAAAEiBAB4AAAAAAAAAUiCABwAAAAAAAIAUCOABAAAAAAAAIAUCeAAAAAAAAABIgQAeAAAAAAAAAFIggAcAAAAAAACAFAjgAQAAAAAAACAFAngAAAAAAAAASIEAHgAAAAAAAABSIIAHAAAAAAAAgBQI4AEAAAAAAAAgBQJ4AAAAAAAAAEiBAB4AAAAAAAAAUiCABwAAAAAAAIAUCOABAAAAAAAAIAUCeAAAAAAAAABIgQAeAAAAAAAAAFIggAcAAAAAAACAFAjgAQAAAAAAACAFAngAAAAAAAAASIEAHgAAAAAAAABSIIAHAAAAAAAAgBQI4AEAAAAAAAAgBQJ4AAAAAAAAAEiBAB4AAAAAAAAAUtAwjZOQjo8++ijMnj07rFixIpSUlISWLVuGrl27hj59+oTGjRvnrK7t27eHt99+O8ydOzesXLky2damTZtw1FFHhV69eoV69erlrDYAAAAAAACAfCGAzwNTp04Nt9xySxJy70yzZs3CoEGDwqhRo0KrVq1qrK4tW7aEu+66K0yYMCEUFxfvdJ/27duHYcOGhaFDh4aioqLUa7j33nvDlVdeucP2xYsXh86dO1fqHP369QuvvvrqHtfwwAMPJJ8/AAAAAAAAQEVMQZ9DmzdvDhdffHE499xzdxm+R+vWrQu/+c1vQvfu3cNrr71WI7UtX748/P3f/3247rrrdhm+R3G0/rXXXhuOPfbYCvfbE/Hcw4cPT/WcAAAAAAAAANVFAJ8j27ZtC+eff3545JFHsrY3aNAgdOnSJfTs2TO0aNEi63erVq0Kp512WnjrrbeqtbY4zXz//v3DX/7yl6ztTZo0CT169AjdunXbYUr8OXPmJMesXr06tTquuuqqsGbNmtTOBwAAAAAAAFCdTEGfI+PHjw9//OMfs7bFqdZvuumm0LZt20xIH/eJU7wvW7Ys2bZhw4YwcODA8N577+0Q0KclTrce16MvFcP2sWPHhiuuuCLsu+++ybb169eHSZMmhRtuuCFs2rQp2fbhhx+Gyy+/PDz99NN7XcOjjz4ann322eTnpk2bJu+XlmnTplVp//jQAQAAAAAAAMDuCOBz4LPPPgu33XZb1rbbb789XH/99Vnb6tevn0xP/53vfCccd9xxYcmSJZmp2e+8884wZsyY1Gt76aWXwgsvvJBpx3XdX3zxxXDCCSdk7RdD8Z/+9KehV69e4eSTT07Wi4+eeeaZMH369GQ0/J6Ko+ivvvrqTPvmm28O11xzTUjLSSedlNq5AAAAAAAAAEqZgj4Hxo0bF9auXZtpx3B7xIgRu9y/Xbt24b777sva9qtf/SoJ8tMWR+CXFR8KKB++l3XiiSfuUPvIkSP3qoahQ4dmprI/5phjssJ4AAAAAAAAgHwlgK9hcVr5Bx54IGvb6NGjQ7169So8bsCAAeH444/PtGOA/8QTT6Ra27vvvhtmz56dNcr9uuuu2+1xw4cPT/YtNXPmzLBgwYI9qiFOO//YY48lPzdo0CD87ne/S/4FAAAAAAAAyHcC+BoWw+lVq1Zl2occckjo169fpY4dPHhwVnvq1Kmp1lZ+Tfq41nzz5s13e1zc57zzztvr2tasWROuuuqqTHvYsGHh6KOPrvJ5AAAAAAAAAHJBAF/Dnnvuuax2XD99d6Pfy+5b1owZM8L69eurrbZTTjml0seWry2OZK+qOJI+rm8fde7cOVn7HQAAAAAAAKC2EMDXsLlz52a1+/TpU+lj27ZtmwTTpUpKSsL8+fNTqWv79u3hnXfe2ePa+vbtm9WeN29ecs7KevXVV8OkSZMy7YkTJ4Z999230scDAAAAAAAA5JoAvoaVXxu9e/fuVTq+/P57utZ6eUuXLg0bNmzItOOa7h07dqz08Z06dcoKzOPI/OXLl1fq2I0bN4YrrrgiE9hfeOGF4dRTTw3V6auvvkoeOHjttdfC22+/nVz/1q1bq/U9AQAAAAAAgMLWMNcF1CUxaF62bFnWtg4dOlTpHOX3/+CDD1Kprfx5qlpX6TFlzxN/rkyIP2rUqPDhhx8mPx9wwAFhwoQJoTrFdeVj+L5t27as7c2aNUtG8v/gBz8Il1xySdhnn32qtQ4AAAAAAACgsAjga9Dq1auzpmUvKioKrVu3rtI52rVrl9VeuXJlKrWVP0/79u2rfI5YW9kAvjK1zZkzJ9x5552Z9vjx46v8meztMgCl1q1bF1588cXk9Ytf/CLcfffd4bzzzquWGuJns2rVqiods3DhwmqpBQAAAAAAAEiHAL4GxYC3rDhle7169ap0jjg1fEXnTKu28u9THbVt2bIlDB48ODP1e79+/cLll18e8sGnn34aBg4cGK699trkoYC03XPPPWHMmDGpnxcAAAAAAADIHWvA16DygXTjxo2rfI4mTZpUeM7aVNsdd9wR5s2bl/wcp3u/9957Q3WJ13PmmWcmwffMmTOTEeglJSVh7dq14aOPPgoPP/xw+Md//McdHoj45S9/GcaOHVttdQEAAAAAAACFwwj4GrRp06asdqNGjap8jvLrksd15WtjbQsWLAi33nprpj1y5Mhw+OGHh+rws5/9LFnb/cADD9zhd3EZgLj2+yGHHBL+6Z/+KbzxxhvhggsuCMXFxZl9brjhhnDaaaeFo446qlrqAwAAAAAAAAqDAL4GlR9VHkdgV9XmzZsrPGdtqG3btm3J1POl+/fo0SOMGDEiVJezzjqr0vsed9xxYcaMGeHYY48Nq1evTrZt3749eUDgmWeeSa2mIUOGVHl9+bgG/DnnnJNaDQAAAAAAAEC6BPA1KI60rmjUeWWUH1Ve/py1obZf//rX4a233kp+jlO+T5o0KRmJni8OO+ywZN33yy67LLPt+eefD59//nk44IADUnmP1q1bJy8AAAAAAACgcFgDvgaVD6Q3bNiQjK6uivXr11d4zrRqK/8+adW2ZMmScOONN2baV155ZejTp0/IN5dcckk46KCDskbtv/zyyzmtCQAAAAAAAMhvAvga1KpVq2TEd6ktW7aElStXVukcZdcmj9IaRV3+PCtWrKjyOSpT2+jRozNBfdu2bcPYsWNDPqpfv37o169f1rYPPvggZ/UAAAAAAAAA+U8AX4OaNGkSOnbsmLVt2bJlVTpH+f27du2aSm1HHHFEVnv58uVVPkf5Y3ZW25dffpn5+eOPPw4tWrRIHkrY3au8Ll26ZP1+woQJIW0dOnTIaq9atSr19wAAAAAAAAAKhwC+hpUPpefPn1+l4xcsWFDh+fZUp06dkgcESsVR6kuXLq308XHfOKV+qaZNm+4QYNc25deljzMWAAAAAAAAAOyKAL6G9ezZM6s9c+bMSh/7ySefJGuolw2Iu3fvnkpdcRT5kUceuce1vfnmm1nteK6djVyvTT799NOsdtk14QEAAAAAAADKa7jDFqrVGWecEe64445M++WXXw7bt2+vVFj90ksvZbX79+8fmjVrlmpts2bNyrSnTZsWLrzwwkodG/ct68wzz9zpfjfffHP4l3/5lyrXdvLJJ2e1H3744dCmTZtdTqGfhjfeeCOrXdtH9AMAAAAAAADVSwBfw/r06RNatWoVVq9enbQXLVoUZsyYkYTpu3P//fdntc8+++xUazvrrLPCTTfdlGlPnjw53H333bsN+deuXZvsW5nayo+y31N9+/YNnTt3DtXl1VdfDR999FHWtgEDBlTb+wEAAAAAAAC1nynoa1j9+vXDoEGDsraNGTMmGQVfkVdeeSW8/vrrmXbz5s3DwIEDU60thuO9e/fOtNetWxfGjRu32+PiPnHN+FLf/e53U5saPxfitQwdOjRr27e//e1wyCGH5KwmAAAAAAAAIP8J4HNgxIgRWaPK42jrstPSl1dcXBx+/OMfZ227+uqrk5H0FYnT2pd9xZH2uxOniC9r7Nix4bXXXtvl/jur/dZbbw35In5OH3/8caX3jzMTxJkA3nnnnR0ekgAAAAAAAACoiAA+B2JwfsMNN2Rt+/nPfx6GDBmSFRZv27YtTJ06NZm2fsmSJZntbdu2Dddcc0211HbqqaeGU045JdPesmVL+N73vhfuuuuusGHDhqxR4hMmTEj2j/uUOv300/NqqvY4hX4cuX7uueeGRx55JOtzLGv58uVh/PjxyUj3P/3pT1m/O+ecc5LjAQAAAAAAACpiDfgcjoKfOXNmePbZZzPbJk6cGCZNmhQ6deoUWrRoERYvXhy+/PLLrOOaNGkSnnjiibD//vtXW20PPfRQOPbYY5P3jzZt2hSGDRuWPCQQw+w4XX5cuz5uL+vQQw8NDz74YMg3mzdvTh5kiK9ov/32C9/4xjeSzzg+PPC3v/1tl6Pkjz/++PDoo4/WcMUAAAAAAABAbSSAz+Fa8JMnTw6XXXZZePzxxzPbt27dmoTbO3PggQeGKVOmhL59+1ZrbW3atAnTp08PZ599dpg3b15m+8aNG8P777+/02N69uwZnn766XDQQQeFfLdmzZrktbv/Ptdee20ynX5RUVGN1QYAAAAAAADUXqagz6HGjRuHxx57LAnVY4C9K02bNk2mp58/f37o169fjdQWR+HPnj07Wd89Tnm/K/F348aNC7NmzQodOnQI+SbOKHDBBRdUuraDDz44WTf+gw8+SK5d+A4AAAAAAABUlhHweeAHP/hB8lq4cGESZBcXF4eSkpJkmvlu3bolI95jWF9Vcar4vdGoUaMwfPjwZCT4nDlzktHwK1euTH7XunXr5KGBXr16JaPFq9ueXssVV1yRvKLPPvssLFiwICxdujSsWrUqWce+QYMGoWXLlqFVq1bh6KOPTqbYBwAAAAAAANgTAvg8cthhhyWvfBMD9t69eyev2ixO4X/cccclLwAAAAAAAIC0mYIeAAAAAAAAAFIggAcAAAAAAACAFAjgAQAAAAAAACAFAngAAAAAAAAASIEAHgAAAAAAAABSIIAHAAAAAAAAgBQI4AEAAAAAAAAgBQJ4AAAAAAAAAEiBAB4AAAAAAAAAUiCABwAAAAAAAIAUCOABAAAAAAAAIAUCeAAAAAAAAABIgQAeAAAAAAAAAFIggAcAAAAAAACAFAjgAQAAAAAAACAFAngAAAAAAAAASIEAHgAAAAAAAABSIIAHAAAAAAAAgBQI4AEAAAAAAAAgBQJ4AAAAAAAAAEiBAB4AAAAAAAAAUiCABwAAAAAAAIAUCOABAAAAAAAAIAUCeAAAAAAAAABIgQAeAAAAAAAAAFIggAcAAAAAAACAFAjgAQAAAAAAACAFAngAAAAAAAAASIEAHgAAAAAAAABSIIAHAAAAAAAAgBQI4AEAAAAAAAAgBQJ4AAAAAAAAAEiBAB4AAAAAAAAAUiCABwAAAAAAAIAUCOABAAAAAAAAIAUCeAAAAAAAAABIgQAeAAAAAAAAAFIggAcAAAAAAACAFAjgAQAAAAAAACAFAngAAAAAAAAASIEAHgAAAAAAAABSIIAHAAAAAAAAgBQI4AEAAAAAAAAgBQJ4AAAAAAAAAEiBAB4AAAAAAAAAUiCABwAAAAAAAIAUCOABAAAAAAAAIAUCeAAAAAAAAABIgQAeAAAAAAAAAFIggAcAAAAAAACAFAjgAQAAAAAAACAFAngAAAAAAAAASIEAHgAAAAAAAABSIIAHAAAAAAAAgBQI4AEAAAAAAAAgBQJ4AAAAAAAAAEiBAB4AAAAAAAAAUiCABwAAAAAAAIAUCOABAAAAAAAAIAUCeAAAAAAAAABIgQAeAAAAAAAAAFIggAcAAAAAAACAFAjgAQAAAAAAACAFAngAAAAAAAAASIEAHgAAAAAAAABSIIAHAAAAAAAAgBQI4AEAAAAAAAAgBQJ4AAAAAAAAAEiBAB4AAAAAAAAAUiCABwAAAAAAAIAUCOABAAAAAAAAIAUCeAAAAAAAAABIgQAeAAAAAAAAAFIggAcAAAAAAACAFAjgAQAAAAAAACAFAngAAAAAAAAASIEAHgAAAAAAAABSIIAHAAAAAAAAgBQ0TOMkpOOjjz4Ks2fPDitWrAglJSWhZcuWoWvXrqFPnz6hcePGOatr+/bt4e233w5z584NK1euTLa1adMmHHXUUaFXr16hXr16obYpxGsCAAAAAAAAcksAnwemTp0abrnlliQQ3plmzZqFQYMGhVGjRoVWrVrVWF1btmwJd911V5gwYUIoLi7e6T7t27cPw4YNC0OHDg1FRUWp13DvvfeGK6+8coftixcvDp07d66V1wQAAAAAAAAUJlPQ59DmzZvDxRdfHM4999xdhu/RunXrwm9+85vQvXv38Nprr9VIbcuXLw9///d/H6677rpdBtVRHK1/7bXXhmOPPbbC/fZEPPfw4cML6poAAAAAAACAwiWAz5Ft27aF888/PzzyyCNZ2xs0aBC6dOkSevbsGVq0aJH1u1WrVoXTTjstvPXWW9VaW5ySvX///uEvf/lL1vYmTZqEHj16hG7duu0wJf6cOXOSY1avXp1aHVdddVVYs2ZNQV0TAAAAAAAAULgE8Dkyfvz48Mc//jFrW5xqfdmyZWHRokVJUPz555+Hp556KnTs2DGzz4YNG8LAgQPDV199VW21xenu43r0pWIwHadsj0H0e++9F+bPn5/8fOedd2aF1h9++GG4/PLLU6nh0UcfDc8++2zyc9OmTQvimgAAAAAAAIDCJoDPgc8++yzcdtttWdtuv/32MHHixNC2bdvMtvr16yfT08+cOTNrvfM4RXoMiqvDSy+9FF544YVMO66B/uKLL4arr7467LvvvpntMRT/6U9/Gv77v/87a530Z555JkyfPn2vaohBeHy/UjfffHOtvyYAAAAAAACg8Angc2DcuHFh7dq1mfYJJ5wQRowYscv927VrF+67776sbb/61a+SID9tN910U1b7+uuvT+rblRNPPHGH2keOHLlXNQwdOjQz7fsxxxyTFcbX1msCAAAAAAAACp8APgdrvz/wwANZ20aPHh3q1atX4XEDBgwIxx9/fKYdA/wnnngi1drefffdMHv27KwR4dddd91ujxs+fHjWNPFxxP6CBQv2qIY47fxjjz2W/NygQYPwu9/9Lvm3Nl8TAAAAAAAAUDcI4GtYDHJXrVqVaR9yyCGhX79+lTp28ODBWe2pU6emWlv5NenjWvPNmzff7XFxn/POO2+va1uzZk246qqrMu1hw4aFo48+OtTmawIAAAAAAADqDgF8DXvuueey2ieffPJuR7+X3besGTNmhPXr11dbbaecckqljy1fWxzJXlVx1Hlc3z6Ka97v7drv+XBNAAAAAAAAQN0hgK9hc+fOzWr36dOn0se2bds2CaZLlZSUhPnz56dS1/bt28M777yzx7X17ds3qz1v3rzknJX16quvhkmTJmXaEydODPvuu2+ozdcEAAAAAAAA1C0C+BpWfh3x7t27V+n48vuntS750qVLw4YNGzLtuP55x44dK318p06dsgLzODJ/+fLllTp248aN4YorrsiE2xdeeGE49dRTQ22+JgAAAAAAAKDuEcDXoBg0L1u2LGtbhw4dqnSO8vt/8MEHqdRW/jxVrWtvahs1alT48MMPk58POOCAMGHChFDbrwkAAAAAAACoexrmuoC6ZPXq1VlTmBcVFYXWrVtX6Rzt2rXLaq9cuTKV2sqfp3379lU+R6ytbEBdmdrmzJkT7rzzzkx7/PjxVf5M8u2aKlvbqlWrqnRM+eUGFi5cmEotdUnJqqW5LqFgvL/P1lyXADtI6//RhPD+++/nuoSCoe9Jj76HfKTvSY++Jz36nvToe8hH+p706HvSo+9Jj76HfKTvSY++Z+8deuihoXHjxiHfCeBr0Lp167LacXrzevXqVekccRr1is6ZVm3l36c6atuyZUsYPHhw2Lr1//+jol+/fuHyyy8PacnFNVXWPffcE8aMGbNX5zjnnHNSqQX2xLdyXQDsxMSJE3NdQsHwWZKP9D3kI/+/TI/Pknyk7yEf+f9lenyW5CN9D/nI/y/T47Pce++9917o0aNHyHemoK9B5cPbPXlCo0mTJhWeszbVdscdd4R58+YlP++zzz7h3nvvDXXl8wYAAAAAAAAKjwC+Bm3atCmr3ahRoyqfIwbV5deVr421LViwINx6662Z9siRI8Phhx8e6srnDQAAAAAAABQeU9DXoPIjsEtKSqp8js2bN1d4ztpQ27Zt25Kp50v3j1NFjBgxItSlz3vIkCHhvPPOq9Ixa9asCX/+85/DfvvtF/bff//QoUOHHR4QAGqPhQsXZi0lMXXq1HDYYYfltCYACpu+B4Capu8BoKbpe6Dw14CvDQTwNahZs2YVjtCujPIjsMufszbU9utf/zq89dZbyc/16tULkyZNCkVFRaEufd6tW7dOXlV17LHHpvL+QP6JNwK1Ye0aAAqHvgeAmqbvAaCm6XuAXDAFfQ0qH95u2LAhbN++vUrnWL9+fYXnTKu28u+TVm1LliwJN954Y6Z95ZVXhj59+oTqUFPXBAAAAAAAABAJ4GtQq1atkhHfpbZs2RJWrlxZpXMUFxdntfdkFPXOlD/PihUrqnyOytQ2evToTKjdtm3bMHbs2FBdauqaAAAAAAAAACIBfA1q0qRJ6NixY9a2ZcuWVekc5ffv2rVrKrUdccQRWe3ly5dX+Rzlj9lZbV9++WXm548//ji0aNEieShhd6/yunTpkvX7CRMm5OyaAAAAAAAAACIBfA0rH+DOnz+/SscvWLCgwvPtqU6dOiUPCJSKo9SXLl1a6ePjvnFK/VJNmzYNHTp0CLlUiNcEAAAAAAAA5C8BfA3r2bNnVnvmzJmVPvaTTz5J1lAvVVRUFLp3755KXXEU+ZFHHrnHtb355ptZ7XiunY1cr0mFeE0AAAAAAABA/mqY6wLqmjPOOCPccccdmfbLL78ctm/fXqlg96WXXspq9+/fPzRr1izV2mbNmpVpT5s2LVx44YWVOjbuW9aZZ5650/1uvvnm8C//8i9Vru3kk0/Oaj/88MOhTZs2u5xuviavCQAAAAAAACASwNewPn36hFatWoXVq1cn7UWLFoUZM2YkYfru3H///Vnts88+O9XazjrrrHDTTTdl2pMnTw533333bkP+tWvXJvtWprbyI9L3VN++fUPnzp3z4poAAAAAAAAAIlPQ17D69euHQYMGZW0bM2ZMMgq+Iq+88kp4/fXXM+3mzZuHgQMHplpbDMd79+6daa9bty6MGzdut8fFfeL66qW++93vpjY1/t4qxGsCAAAAAAAA8pMAPgdGjBiRNQL71VdfzZqWvrzi4uLw4x//OGvb1VdfnYykr0ic1r7sK4603504RXxZY8eODa+99tou999Z7bfeemvIJ4V4TQAAAAAAAED+EcDnQAzOb7jhhqxtP//5z8OQIUPCxx9/nNm2bdu2MHXq1GTa+iVLlmS2t23bNlxzzTXVUtupp54aTjnllEx7y5Yt4Xvf+1646667woYNGzLb4+jwCRMmJPvHfUqdfvrpYcCAASGfFOI1AQAAAAAAAPlHAJ/DUfBnnHFG1raJEyeGjh07hkMPPTT06tUrHHjggeHcc88Ny5Yty+zTpEmT8MQTT4T999+/2mp76KGHQpcuXTLtTZs2hWHDhiUPDnzrW98KPXr0SH7+6U9/mvyuVKz7wQcfDPmoEK8JAAAAAAAAyC8Nc11AXV4LfvLkyeGyyy4Ljz/+eGb71q1bw6JFi3Z6TAzkp0yZEvr27VuttbVp0yZMnz49nH322WHevHmZ7Rs3bgzvv//+To/p2bNnePrpp8NBBx0U8lEhXhNQu8X/t4waNSqrDQDVSd8DQE3T9wBQ0/Q9QD6ot3379u25LqKue/LJJ5M1xufOnbvT3zdt2jRceumlSafRunXrSp83rvteVgyg+/XrV+njS0pKkinZ41TtZafGLytOhx9Hksc16Rs1ahSqS/lrWbx4cejcuXOVz5NP1wQAAAAAAAAUFgF8Hlm4cGGYNWtWKC4uToLiOM18t27dkhHvjRs3zlldcS36OXPmJCPHV65cmWyLDwLEEeJxqvw4mr+2KcRrAgAAAAAAAHJLAA8AAAAAAAAAKTDMFwAAAAAAAABSIIAHAAAAAAAAgBQI4AEAAAAAAAAgBQJ4AAAAAAAAAEiBAB4AAAAAAAAAUiCABwAAAAAAAIAUCOABAAAAAAAAIAUCeAAAAAAAAABIgQAeAAAAAAAAAFIggAcAAAAAAACAFAjgAQAAAAAAACAFAngAAAAAAAAASEHDNE4CAAAA1e2jjz4Ks2fPDitWrAglJSWhZcuWoWvXrqFPnz6hcePGuS4PAAAoMNu3bw9LliwJ7777bnIf8uWXX4Z99tknuRf55je/GXr37l0t9yLufaB2E8ADQB0yevToMGbMmD0+/tJLLw0PPvhgqjUBUPsUFxcnXwbNmjUr+ffPf/5zWLt2beb3nTp1Sr6kSsvUqVPDLbfcEt5+++2d/r5Zs2Zh0KBBYdSoUaFVq1apvS8Adav/cb8EQPTFF18k9yD//d//Hf70pz+F1atX73LfoqKi8I//+I9h2LBh4cQTT9zr93bvA4VBAA8AAMBuvfnmm+Hf//3fk9Dj448/rpH33Lx5cxg8eHB45JFHKtxv3bp14Te/+U34wx/+EKZMmRJOOOGEGqkPgMLsfwCou37yk5+E++67Lxl1XhlbtmxJQvP4uuSSS8Kvf/3rsN9++1X5fd37QGGxBjwAAAC79f/+3/8L//Vf/1Vj4ce2bdvC+eefv8MXUA0aNAhdunQJPXv2DC1atMj63apVq8Jpp50W3nrrrRqpEYDC638AqNviA187C9/jfUj79u3DMcccE4488sgd7kWihx56KJx88slJSF4V7n2g8BgBDwB12C9/+ctw1FFHVXr/tm3bVms9ANROcRrEqn7JtDvjx48Pf/zjH7O2XXnlleGmm27K9Efxi6q4T5zucdmyZcm2DRs2hIEDB4b33ntvp1+KAVA4qqP/Kcv9EkDdtv/++4eLLroomWL++OOPD82bN8/8buvWreH1118Pv/jFL5J/S8UlUuIU8XF0emW594HCI4AHgDosPrXbr1+/XJcBQC0Sv3SK/Ufv3r3Dd77zneTfxYsXh/79+6f2Hp999lm47bbbsrbdfvvt4frrr8/aVr9+/XDuuecmdRx33HGZdX9XrFgR7rzzzr1axxeAutf/lOd+CaBu6ty5cxg5cmQSvjdp0mSn+8TR6bGPmD59ehgyZEiYNGlS5ndPPvlksr0yfZR7HyhMAngAAAB268wzzwynnHJK6Nq1a/LlT1kxAEnTuHHjwtq1azPtuK7hiBEjdrl/u3btknUaTzrppMy2X/3qV2Ho0KHhwAMPTLU2AAq3/wGAGGTHaeQbNWpUqf1jEH/PPfeEt99+O/z5z3/ObI/3J5UJ4N37QGGyBjwAAAC7deihh4bu3bvvEH6kLU6t+MADD2RtGz16dKhXr16Fxw0YMCCZFrJU/BLriSeeqLY6ASis/gcAojjdfGXD97Ih/PDhw7O2vfjii7s9zr0PFC4j4AEoePGP2Q8++CDMmzcvrF69Onz11VehYcOGyRSGHTt2DEcccUT45je/mesyAchD+pCaN3PmzLBq1apM+5BDDqn09L+DBw/OWn9x6tSp4aqrrqqWOgGqk/4HgJqm79k7ZQPx0qnl4xrt++677y6Pce8DhUsAD0DBWrhwYbj77rvDo48+mvzRW5E4RdM//MM/hB/96EfJFIcA1G36kNx57rnnstpx+sfdjQApu29ZM2bMCOvXrw9NmzZNtUaA6qL/AaCm6XvS0bJlyx22xYcYKgrg3ftA4RLAA1CQbr/99mTNps2bN1dq/3iDMXny5ORnNxAAdZs+JLfmzp2b1e7Tp0+lj23btm3o3LlzWLJkSdIuKSkJ8+fPD7179069ToC06X8AqGn6nvQUFxfvsG13a7K794HCJYAHoOD85Cc/Cffcc89On0SNU2Y1adIkeSJ05cqV4W9/+1vWPr169Qp1TbzJWrRoUXITVVRUlNwcxD/iK3pCF6BQ6UNyb8GCBVntuO5vVcT9S7+EKj2fL6GAfKf/yV/ul4BCpe9JV9np4KNOnTrtdi159z5QuATwABSUP/zhD1k3D/vtt1/42c9+Fv7pn/4pHHbYYTvsH28i4hRNTz31VPI6+uijQ1272YpfJm3atClre1zj65hjjgmnnXZaGDJkSDjooINyViNATdGH5N7GjRvDsmXLsrZ16NChSucov39cxxIgn+l/8pf7JaBQ6XvS9x//8R9Z7dNPP73C/d37QGGrn+sCACBNI0eOzPzcpk2bMGfOnDBq1Kid3jxErVu3DgMHDgyPP/54+Oijj6o01VOpQYMGJeszVffrwQcfDGmLU1OV/zIp+vrrr8OsWbPC6NGjkyd2f/GLX4StW7em/v4A+UQfknurV68O27dvz7TjSMP4OVdFu3btdviyECCf6X/yl/sloFDpe9L1/PPPh9dee22H662Iex8obAJ4AArGvHnzwsKFCzPtG264YZc3Drt6arR58+bVVF3tFZ/IveWWW8JJJ50U1q1bl+tyAKqFPiQ/lO9n4vS+8Uu0qmjatGmF5wTIJ/qf2s/9ElDb6HvS9fnnn4d//ud/ztp2zjnnhO985zsVHufeBwqbAB6AglF+mqXDDz88Z7Xks/jHfHxS+bbbbgvTpk0LK1asCBs2bEhGdhQXF4dnnnkmuXFo3Lhx1nFxqrELLrjAyA6gIOlD8kP5L4zK90WVEdeqrOicAPlE/5N/3C8BhU7fk55t27aFiy++OOkrSrVo0SLcfffduz3WvQ8UNmvAA1Awvvrqq6z21KlTw6mnnlrt7zt8+PDkj+3q1qNHj70+xymnnBIuuuiiXd5ctW3bNnmdccYZyXRk8QukN998M/P75557Llkj7F//9V/3uhaAfKIPyQ/lp/lt1KhRlc+xzz777DAyESBf6X/yi/sloC7Q96TnuuuuCy+88ELWtnvvvbdSa7m794HCJoAHoGB07tx5hz94GzRoEG699dbQsmXLanvf7t27J6/aoCprdLVv3z68/PLL4R/+4R/CW2+9ldkeP8/BgwcnU2MBFAp9SH4oP+qjpKSkyufYvHlzhecEyCf6n/zifgmoC/Q96Yij3O+8884dHjI4//zzK3W8ex8obKagB6Bg9OvXLxx66KFZ2+Log/jU6YUXXhimTJkS1qxZk7P6aqP4h/tDDz0UGjb8v2f2Vq5cGV566aWc1gWQNn1IfmjWrFmFo0Iqo/yoj/LnBMgn+p/azf0SUBvpe/beo48+GoYNG5a1bdCgQWHs2LGVPod7HyhsAngACkZRUVF48sknw8EHH5y1ff369eHxxx8P5513XmjVqlU4+eSTkxuLzz//PGe11iaHHXZYOOuss7K2+UIJKDT6kPxQ/gujuObu9u3bq3SO+N+sonMC5BP9T+3nfgmobfQ9e+fZZ58Nl156adZ9yve///1w3333hXr16lX6PO59oLAJ4AEoKEcddVR45513wtChQ3f6R+eWLVuSaQJ/8pOfJGv3XXnllWHVqlU5qbU2GTBgQFb7gw8+yFktANVFH5J78Yu+sl9axc88jiSsiuLi4qx269atU6sPoDrof2o/90tAbaPv2TPTp09PHlD4+uuvM9vigwqPPfZYMo1/Vbj3gcJWb3tVH6kBgFpi3bp14YUXXggvvvhi8gfyokWLdrpfvJF44403QpcuXfbofebPnx8+/vjjUN169OgRvvGNb4RceOaZZ7JGdXz7299ObtQACpU+pGpmzJgR+vfvn2l36tQpLFmyZI/XpFy6dGmmPXv27NC7d+9KH3/66acn/+1K/ed//me45JJL9qgWgJqm/8ld/7M33C8BtZm+p3JmzZoVTjrppOTzKtWnT59k1pOmTZvu0Tnd+0Dh+r8FigCgwMQneONTqfEV/fWvf02eSP3tb38bPv3008x+8Y//a665Jjz11FN79D7jxo1L/sCtbg888ECynlSupicrKz6VC1DI9CG507Vr16wvoeIXdVX5EmrBggU7nA+gttD/1E7ul4DaTN+ze/GhqtNOOy0rfD/66KPD888/v8fhe+TeBwqXKegBqDMOP/zwMGrUqLBw4cLwox/9KOt38WlRk8LsWtkbruiggw7KWS0AuaAPqTk9e/bMas+cObPSx37yySdZIx9jINK9e/dU6wOoSfqf2sH9ElBI9D1hh2VF4jTzX3zxRWZbt27dkhkDWrRosVfndu8DhUsAD0CdE59MjU/DduzYMbNt06ZNYfPmzTmtK5/FKcbK6tChQ85qAcglfUj1O+OMM7Lace3Jyn7JF6d/LCtOS7yzNS0Bahv9T35zvwQUIn1PSEanx2nny67NHqfgnzZtWioPW7n3gcIlgAegTmrQoEE48MADM+34xGrjxo336FwPPvhg8sdxdb9yNXXjl19+GZ588smsbQMGDMhJLQD5QB9SveI6iq1atcq04xqUcY3fyrj//vuz2meffXbq9QHkiv4nP7lfAgpZXe574gjz+P/zFStWZLa1a9cuvPLKK8m/aXDvA4VLAA9AnbRs2bJk/aZSVVlfqa659tprky+VSjVq1ChZ9wqgrtKHVK/69evv8KXZmDFjdjsSJH4R9vrrr2fazZs3DwMHDqy2OgFqmv4nP7lfAgpZXe17Pv/882Ta+Y8++iizLY54jyPf4wj4tLj3gcIlgAeg1rvwwgvDf/3Xf4Vt27ZVav/45cgFF1wQtm7dmtl2ySWXhEI3duzYMGfOnErv//XXX4drrrlmhydqr7zyyvCNb3yjGioEqHn6kPw0YsSIrOkTX3311XDHHXfscv/i4uLw4x//OGvb1VdfnTWaBCCf6H/yj/sloNDpeypn7dq14dRTTw3vv/9+Ztv++++fTPke135Pm3sfKEz1tld2QQkAyEPxZqBly5bJz+3btw8//OEPw5lnnpk8kRuf/izriy++CJMnTw633nprWL58eWZ73759k6dG69WrFwpZv379kj/i4/RW8anYOI1W165dQ8OGDbP2++qrr8Lzzz8fxo0bF+bOnZv1u0MPPTTMmjUra/oxgNpKH1J1b775Zti4ceMO2+fNm5eMACzVpk2b8PDDD+/0HG3btg3du3ff7Xvdfvvt4YYbbsjadtVVV4WRI0cm54jil4dPP/108oVTHJ1T9j3iF2bxizKAfKP/yc/+x/0SUMj0PZUX11IvPw38zTffHI499tgqn+uYY47JfO4Vce8DhUcAD0CtFv8gjn8Y72wKp3hDEb/4KCoqCqtWrQpLlizZYQqn+IVKnLap9I/ZQlb6hVJZ++yzT/I5xTW84rpen332WfI57exp6IMPPji89tpr4Zvf/GYNVg1QffQhVde5c+ewdOnSvTrHpZdemqz/uDuxL4rrGD777LNZ22N/1alTp6TvWrx4cda0v1GTJk2SqSHjF4QA+Uj/k5/9j/sloJDpeyovzQcMpk+fnvQvu+PeBwpP9iOcAFDLlB9xUPYP1/g0aNknQsv7/ve/H377298mazjVVZs3b85az2pXTj/99PDAAw+E1q1b10hdADVBH5Lf4peBceTNZZddFh5//PHM9jgF5qJFi3Z6TPzicMqUKb6AAvKa/qf2cL8EFAp9T35z7wOFxxrwANRqcc2jp556KlxxxRXhiCOO2O1TqvFp3nPOOSd5OvTJJ5+sUzcPN954Y7IeYY8ePZInaHcnrj913nnnJaNAnnvuOV8mAQVHH5L/GjduHB577LHki6WePXvucr+mTZuGIUOGhPnz51dqhAlALul/8pP7JaCQ6Xvyn3sfKCymoAegoMR1quIfoHGUwurVq8OGDRuS6ZjiektxuqxevXolf9DWdfFziZ9TnFbsk08+CevWrUueeo7rRcXPKq6N+O1vf7tSXzwBFAp9SP5buHBhsrZucXFxKCkpSfqtbt26JaM+/LcBaiv9T/5xvwQUOn1P/nPvA7WbAB4AAAAAAAAAUmAKegAAAAAAAABIgQAeAAAAAAAAAFIggAcAAAAAAACAFAjgAQAAAAAAACAFAngAAAAAAAAASIEAHgAAAAAAAABSIIAHAAAAAAAAgBQI4AEAAAAAAAAgBQJ4AAAAAAAAAEiBAB4AAAAAAAAAUiCABwAAAAAAAIAUCOABAAAAAAAAIAUCeAAAAAAAAABIgQAeAAAAAAAAAFIggAcAAAAAAACAFAjgAQAAAAAAACAFAngAAAAAAAAASIEAHgAAAAAAAABSIIAHAAAAAAAAgBQI4AEAAAAAAAAgBQJ4AAAAAAAAAEiBAB4AAAAAAAAAUiCABwAAAAAAAIAUCOABAAAAAAAAIAUCeAAAAAAAAABIgQAeAAAAAAAAAFIggAcAAAAAAACAFAjgAQAAAAAAACAFAngAAAAAAAAASIEAHgAAAAAAAABSIIAHAAAAAAAAgBQI4AEAAAAAAAAgBQJ4AAAAAAAAAEiBAB4AAAAAAAAAUiCABwAAAAAAAIAUCOABAAAAAAAAIOy9/w9HbQ/bgUbFNgAAAABJRU5ErkJggg==",
      "text/plain": [
       "<Figure size 2100x900 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import os\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "from scipy.stats import mannwhitneyu, sem\n",
    "\n",
    "# Configuration\n",
    "RESULTS_ROOT = os.path.abspath(\n",
    "    os.path.join(\n",
    "        os.getcwd(), os.pardir, os.pardir, \"results\", \"experiment_1\"\n",
    "    )\n",
    ")\n",
    "fish_list = [9, 10, 11, 12, 13]\n",
    "seq_lengths = [5, 10, 15, 20]\n",
    "num_runs = 10\n",
    "\n",
    "model_map = {\n",
    "    \"GPT-2 (base)\":      (\"gpt2_pretrained\", \"#1f77b4\"),\n",
    "    \"LSTM\":              (\"lstm\",            \"#ff7f0e\"),\n",
    "    \"DeepSeek-coder-7b\": (\"deepseek_moe\",    \"#000000\"),\n",
    "    \"BERT-small\":        (\"bert\",            \"#7f7f7f\"),\n",
    "}\n",
    "model_names = list(model_map.keys())\n",
    "print(\"Models used:\", \", \".join(model_names))\n",
    "\n",
    "# Prepare data structure\n",
    "rmse_data = {m: {s: [] for s in seq_lengths} for m in model_names}\n",
    "\n",
    "# Load and compute RMSEs\n",
    "for fish in fish_list:\n",
    "    gt_path = os.path.join(\n",
    "        RESULTS_ROOT,\n",
    "        f\"fish{fish}\",\n",
    "        f\"fish{fish}_final_predictions_groundtruth_test.npy\"\n",
    "    )\n",
    "    y_true = np.load(gt_path)\n",
    "    for run in range(1, num_runs + 1):\n",
    "        for seq in seq_lengths:\n",
    "            seq_dir = os.path.join(\n",
    "                RESULTS_ROOT,\n",
    "                f\"fish{fish}\",\n",
    "                f\"run_{run}\",\n",
    "                f\"seq_{seq}\"\n",
    "            )\n",
    "            if not os.path.isdir(seq_dir):\n",
    "                continue\n",
    "            for m in model_names:\n",
    "                key = model_map[m][0]\n",
    "                fpath = os.path.join(\n",
    "                    seq_dir,\n",
    "                    f\"fish{fish}_final_predictions_{key}_test_run{run}.npy\"\n",
    "                )\n",
    "                if os.path.isfile(fpath):\n",
    "                    preds = np.load(fpath)\n",
    "                    rmse = np.sqrt(np.mean((preds - y_true) ** 2))\n",
    "                    rmse_data[m][seq].append(rmse)\n",
    "\n",
    "# Compute means, errors, and statistical tests\n",
    "means = {}\n",
    "errs = {}\n",
    "pvals = {}\n",
    "for m in model_names:\n",
    "    means[m] = [np.mean(rmse_data[m][s]) for s in seq_lengths]\n",
    "    errs[m] = [sem(rmse_data[m][s]) for s in seq_lengths]\n",
    "\n",
    "for seq in seq_lengths:\n",
    "    lstm_vals = rmse_data[\"LSTM\"][seq]\n",
    "    pvals[seq] = {}\n",
    "    for other in [\"BERT-small\", \"DeepSeek-coder-7b\"]:\n",
    "        stat, p = mannwhitneyu(\n",
    "            lstm_vals,\n",
    "            rmse_data[other][seq],\n",
    "            alternative=\"two-sided\"\n",
    "        )\n",
    "        pvals[seq][other] = p\n",
    "\n",
    "# Bar plot parameters\n",
    "gap = 0.0005\n",
    "h = 0.0008\n",
    "y_bases = [\n",
    "    max(means[m][j] + errs[m][j] for m in model_names) + gap\n",
    "    for j in range(len(seq_lengths))\n",
    "]\n",
    "\n",
    "plt.rcParams[\"pdf.fonttype\"] = 42\n",
    "fig, ax = plt.subplots(figsize=(7, 3), dpi=300)\n",
    "\n",
    "x = np.arange(len(seq_lengths))\n",
    "bw = 0.8 / len(model_names)\n",
    "\n",
    "# Plot bars\n",
    "for idx, m in enumerate(model_names):\n",
    "    off = (idx - len(model_names) / 2) * bw + bw / 2\n",
    "    ax.bar(\n",
    "        x + off,\n",
    "        means[m],\n",
    "        width=bw,\n",
    "        yerr=errs[m],\n",
    "        capsize=3,\n",
    "        label=m,\n",
    "        color=model_map[m][1]\n",
    "    )\n",
    "\n",
    "ax.set_ylim(0.04, None)\n",
    "ax.set_ylabel(\"RMSE\")\n",
    "ax.set_xticks(x)\n",
    "ax.set_xticklabels([f\"$s={s}$\" for s in seq_lengths])\n",
    "ax.spines[\"top\"].set_visible(False)\n",
    "ax.spines[\"right\"].set_visible(False)\n",
    "\n",
    "ax.legend(\n",
    "    ncol=len(model_names),\n",
    "    frameon=False,\n",
    "    bbox_to_anchor=(0, 1.12, 1, 0),\n",
    "    loc=\"lower center\"\n",
    ")\n",
    "\n",
    "# Function to draw significance brackets\n",
    "def draw_bracket(x1, x2, y, h, p):\n",
    "    ax.plot([x1, x1, x2, x2], [y, y + h, y + h, y], lw=1, c=\"k\")\n",
    "    if p < 0.05:\n",
    "        ax.text(\n",
    "            (x1 + x2) / 2,\n",
    "            y + h + 0.001,\n",
    "            \"*\",\n",
    "            ha=\"center\",\n",
    "            va=\"bottom\",\n",
    "            fontsize=10\n",
    "        )\n",
    "\n",
    "# Add brackets\n",
    "for j, seq in enumerate(seq_lengths):\n",
    "    y_base = y_bases[j]\n",
    "    idx_l = model_names.index(\"LSTM\")\n",
    "    idx_b = model_names.index(\"BERT-small\")\n",
    "    idx_d = model_names.index(\"DeepSeek-coder-7b\")\n",
    "\n",
    "    x_l = x[j] + ((idx_l - len(model_names) / 2) * bw + bw / 2)\n",
    "    x_b = x[j] + ((idx_b - len(model_names) / 2) * bw + bw / 2)\n",
    "    x_d = x[j] + ((idx_d - len(model_names) / 2) * bw + bw / 2)\n",
    "\n",
    "    draw_bracket(x_l, x_b, y_base, h, pvals[seq][\"BERT-small\"])\n",
    "    draw_bracket(x_l, x_d, y_base + h + 0.0003, h, pvals[seq][\"DeepSeek-coder-7b\"])\n",
    "\n",
    "plt.tight_layout()\n",
    "out_file = os.path.join(RESULTS_ROOT, \"experiment1_rmse_sig_final.pdf\")\n",
    "plt.savefig(out_file, dpi=300, bbox_inches=\"tight\")\n",
    "print(\"Saved to\", out_file)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "cc7fc81d-fcf0-4dfd-834d-7f91b71df429",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\jacob\\AppData\\Local\\Temp\\ipykernel_22560\\3963392535.py:170: UserWarning: This figure includes Axes that are not compatible with tight_layout, so results might be incorrect.\n",
      "  plt.tight_layout()\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved to C:\\Users\\jacob\\OneDrive\\Documents\\GitHub\\external_data\\nlp4neuro\\results_april30\\experiment_1\\experiment1_rmse_sig_final.pdf\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 2100x1020 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import os\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "from scipy.stats import mannwhitneyu, sem\n",
    "\n",
    "RESULTS_ROOT = os.path.abspath(\n",
    "    os.path.join(\n",
    "        os.getcwd(), os.pardir, os.pardir, \"results\", \"experiment_1\"\n",
    "    )\n",
    ")\n",
    "fish_list = [9, 10, 11, 12, 13]\n",
    "seq_lengths = [5, 10, 15, 20]\n",
    "num_runs = 10\n",
    "\n",
    "model_map = {\n",
    "    \"GPT-2 (base)\":      (\"gpt2_pretrained\", \"#1f77b4\", \"GPT-2\"),\n",
    "    \"LSTM\":              (\"lstm\",            \"#ff7f0e\", \"LSTM\"),\n",
    "    \"DeepSeek-coder-7b\": (\"deepseek_moe\",    \"#000000\", \"DeepSeek-c7b\"),\n",
    "    \"BERT-small\":        (\"bert\",            \"#7f7f7f\", \"BERT-bu\"),\n",
    "    \"RC\":                (\"reservoir\",       \"purple\",  \"RC\"),\n",
    "}\n",
    "model_names = list(model_map.keys())\n",
    "\n",
    "# gather RMSEs\n",
    "rmse_data = {m: {s: [] for s in seq_lengths} for m in model_names}\n",
    "for fish in fish_list:\n",
    "    y_true = np.load(os.path.join(\n",
    "        RESULTS_ROOT,\n",
    "        f\"fish{fish}\",\n",
    "        f\"fish{fish}_final_predictions_groundtruth_test.npy\"\n",
    "    ))\n",
    "    for run in range(1, num_runs + 1):\n",
    "        for s in seq_lengths:\n",
    "            seq_dir = os.path.join(RESULTS_ROOT, f\"fish{fish}\", f\"run_{run}\", f\"seq_{s}\")\n",
    "            if not os.path.isdir(seq_dir):\n",
    "                continue\n",
    "            for m in model_names:\n",
    "                key = model_map[m][0]\n",
    "                fp = os.path.join(\n",
    "                    seq_dir,\n",
    "                    f\"fish{fish}_final_predictions_{key}_test_run{run}.npy\"\n",
    "                )\n",
    "                if os.path.isfile(fp):\n",
    "                    preds = np.load(fp)\n",
    "                    rmse = np.sqrt(np.mean((preds - y_true) ** 2))\n",
    "                    rmse_data[m][s].append(rmse)\n",
    "\n",
    "# compute means, errors, and p-values\n",
    "means, errs, pvals = {}, {}, {}\n",
    "for m in model_names:\n",
    "    means[m] = [np.mean(rmse_data[m][s]) for s in seq_lengths]\n",
    "    errs[m]  = [sem(rmse_data[m][s])       for s in seq_lengths]\n",
    "\n",
    "for s in seq_lengths:\n",
    "    lst = rmse_data[\"LSTM\"][s]\n",
    "    pvals[s] = {\n",
    "        \"BERT-small\":        mannwhitneyu(lst, rmse_data[\"BERT-small\"][s])[1],\n",
    "        \"DeepSeek-coder-7b\": mannwhitneyu(lst, rmse_data[\"DeepSeek-coder-7b\"][s])[1],\n",
    "    }\n",
    "\n",
    "# plot with broken y-axis\n",
    "plt.rcParams[\"pdf.fonttype\"] = 42\n",
    "fig, (ax_hi, ax_lo) = plt.subplots(\n",
    "    2, 1, sharex=True,\n",
    "    figsize=(7, 3.4), dpi=300,\n",
    "    gridspec_kw={\"height_ratios\":[1,2], \"hspace\":0.05}\n",
    ")\n",
    "\n",
    "x  = np.arange(len(seq_lengths))\n",
    "bw = 0.8 / len(model_names)\n",
    "\n",
    "for i, m in enumerate(model_names):\n",
    "    off = (i - len(model_names)/2) * bw + bw/2\n",
    "    col, lbl = model_map[m][1], model_map[m][2]\n",
    "\n",
    "    ax_lo.bar(\n",
    "        x + off,\n",
    "        means[m],\n",
    "        width = bw,\n",
    "        yerr  = None if m==\"RC\" else errs[m],\n",
    "        capsize = 0 if m==\"RC\" else 3,\n",
    "        color = col,\n",
    "        label = lbl\n",
    "    )\n",
    "    if m == \"RC\":\n",
    "        ax_hi.bar(\n",
    "            x + off,\n",
    "            means[m],\n",
    "            width = bw,\n",
    "            yerr  = errs[m],\n",
    "            capsize = 3,\n",
    "            color = col,\n",
    "            label = \"_nolegend_\"\n",
    "        )\n",
    "\n",
    "ax_lo.set_ylim(0.045, 0.065)\n",
    "ax_hi.set_ylim(0.23, max(means[\"RC\"]) + errs[\"RC\"][-1] + 0.01)\n",
    "\n",
    "for ax in (ax_hi, ax_lo):\n",
    "    ax.spines[\"right\"].set_visible(False)\n",
    "    ax.spines[\"top\"].set_visible(False)\n",
    "ax_hi.spines[\"bottom\"].set_visible(False)\n",
    "ax_hi.tick_params(axis='x', bottom=False, labelbottom=False)\n",
    "\n",
    "d = 0.015\n",
    "ax_lo.plot((-d, +d), (1-d, 1+d), transform=ax_lo.transAxes, color=\"k\", clip_on=False)\n",
    "ax_hi.plot((-d, +d), (-2*d, +2*d), transform=ax_hi.transAxes, color=\"k\", clip_on=False)\n",
    "\n",
    "ax_lo.set_ylabel(\"RMSE\")\n",
    "ax_lo.set_xticks(x)\n",
    "ax_lo.set_xticklabels([f\"$s={s}$\" for s in seq_lengths])\n",
    "\n",
    "# significance bars\n",
    "sig_gap      = 0.0005\n",
    "sig_hgt      = 0.0008\n",
    "sig_star_off = 0.001\n",
    "sig_between  = 0.0003\n",
    "sig_overrides = {\n",
    "    5: dict(gap=0.0002, h=0.0005, star_off=0.0005)\n",
    "}\n",
    "\n",
    "def draw_bracket(ax, x1, x2, y, h, p, star_off=sig_star_off, always=False):\n",
    "    if always or p < 0.05:\n",
    "        ax.plot([x1, x1, x2, x2], [y, y+h, y+h, y], lw=1, c=\"k\")\n",
    "        ax.text(\n",
    "            (x1+x2)/2, y+h+star_off,\n",
    "            \"*\", ha=\"center\", va=\"bottom\", fontsize=10\n",
    "        )\n",
    "\n",
    "iL = model_names.index(\"LSTM\")\n",
    "iB = model_names.index(\"BERT-small\")\n",
    "iD = model_names.index(\"DeepSeek-coder-7b\")\n",
    "\n",
    "for j, s in enumerate(seq_lengths):\n",
    "    params = sig_overrides.get(s, dict(gap=sig_gap, h=sig_hgt, star_off=sig_star_off))\n",
    "    y0 = max(\n",
    "        means[m][j] + errs[m][j]\n",
    "        for m in model_names if m != \"RC\"\n",
    "    ) + params[\"gap\"]\n",
    "    y0 = min(y0, ax_lo.get_ylim()[1] - params[\"h\"] - 0.0002)\n",
    "\n",
    "    xL = x[j] + ((iL - len(model_names)/2) * bw + bw/2)\n",
    "    xB = x[j] + ((iB - len(model_names)/2) * bw + bw/2)\n",
    "    xD = x[j] + ((iD - len(model_names)/2) * bw + bw/2)\n",
    "\n",
    "    if s == 5:\n",
    "        draw_bracket(\n",
    "            ax_lo, xL, xB, y0,\n",
    "            params[\"h\"], pvals[s][\"BERT-small\"],\n",
    "            star_off=params[\"star_off\"], always=True\n",
    "        )\n",
    "    else:\n",
    "        draw_bracket(ax_lo, xL, xB, y0, params[\"h\"], pvals[s][\"BERT-small\"])\n",
    "        draw_bracket(\n",
    "            ax_lo, xL, xD,\n",
    "            y0 + params[\"h\"] + sig_between,\n",
    "            params[\"h\"], pvals[s][\"DeepSeek-coder-7b\"]\n",
    "        )\n",
    "\n",
    "ax_lo.legend(\n",
    "    ncol=len(model_names), frameon=False,\n",
    "    loc=\"upper center\", bbox_to_anchor=(0.5, -0.18)\n",
    ")\n",
    "plt.subplots_adjust(bottom=0.22)\n",
    "\n",
    "out_file = os.path.join(RESULTS_ROOT, \"experiment1_rmse_sig_final.pdf\")\n",
    "plt.tight_layout()\n",
    "plt.savefig(out_file, bbox_inches=\"tight\")\n",
    "print(\"Saved to\", out_file)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a0826f41-4288-4f68-9316-d5bf0cb2b891",
   "metadata": {},
   "source": [
    "## 2) Plot pre- vs un-trained"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "id": "f4d457b6-c0e9-4288-aab4-3b831629a493",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved to C:\\Users\\jacob\\OneDrive\\Documents\\GitHub\\external_data\\nlp4neuro\\results_april30\\experiment_2\\pretrained_vs_untrained_exp1_exp2.pdf\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 2100x900 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import os\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "from scipy.stats import mannwhitneyu, sem, wilcoxon\n",
    "import matplotlib.patches as mpatches\n",
    "\n",
    "EXP1_ROOT = os.path.abspath(\n",
    "    os.path.join(\n",
    "        os.getcwd(), os.pardir, os.pardir, \"results\", \"experiment_1\"\n",
    "    )\n",
    ")\n",
    "EXP2_ROOT = os.path.abspath(\n",
    "    os.path.join(\n",
    "        os.getcwd(), os.pardir, os.pardir, \"results\", \"experiment_2\"\n",
    "    )\n",
    ")\n",
    "\n",
    "fish_list = [9, 10, 11, 12, 13]\n",
    "seq_lengths = [5, 20]\n",
    "model_info = {\n",
    "    \"GPT-2 (base)\":      {\"pre_key\": \"gpt2_pretrained\", \"stub\": \"GPT2\",     \"c\": \"#1f77b4\"},\n",
    "    \"DeepSeek-coder-7b\": {\"pre_key\": \"deepseek_moe\",    \"stub\": \"DeepSeek\", \"c\": \"#000000\"},\n",
    "    \"BERT-small\":        {\"pre_key\": \"bert\",            \"stub\": \"BERT\",     \"c\": \"#7f7f7f\"},\n",
    "}\n",
    "states = {\"prtr\": \"Pretrained\", \"untr\": \"Untrained\"}\n",
    "\n",
    "def load_gt(fid):\n",
    "    for root in (EXP1_ROOT, EXP2_ROOT):\n",
    "        for cand in (f\"fish{fid}_test_groundtruth.npy\", f\"fish{fid}_groundtruth_test.npy\"):\n",
    "            fp = os.path.join(root, cand)\n",
    "            if os.path.isfile(fp):\n",
    "                return np.load(fp)\n",
    "        sub = os.path.join(root, f\"fish{fid}\")\n",
    "        for cand in (f\"fish{fid}_test_groundtruth.npy\", f\"fish{fid}_groundtruth_test.npy\"):\n",
    "            fp = os.path.join(sub, cand)\n",
    "            if os.path.isfile(fp):\n",
    "                return np.load(fp)\n",
    "    raise FileNotFoundError(f\"GT fish{fid}\")\n",
    "\n",
    "# collect RMSEs\n",
    "rmse = {s: {m: {\"prtr\": [], \"untr\": []} for m in model_info} for s in seq_lengths}\n",
    "for fish in fish_list:\n",
    "    y_true = load_gt(fish)\n",
    "    for run in range(1, 11):\n",
    "        for s in seq_lengths:\n",
    "            seq_dir = os.path.join(EXP1_ROOT, f\"fish{fish}\", f\"run_{run}\", f\"seq_{s}\")\n",
    "            if not os.path.isdir(seq_dir):\n",
    "                continue\n",
    "            for m, meta in model_info.items():\n",
    "                fp = os.path.join(\n",
    "                    seq_dir,\n",
    "                    f\"fish{fish}_final_predictions_{meta['pre_key']}_test_run{run}.npy\"\n",
    "                )\n",
    "                if os.path.isfile(fp):\n",
    "                    preds = np.load(fp)\n",
    "                    rmse_val = np.sqrt(np.mean((preds - y_true) ** 2))\n",
    "                    rmse[s][m][\"prtr\"].append(rmse_val)\n",
    "    for run in range(1, 11):\n",
    "        for s in seq_lengths:\n",
    "            for m, meta in model_info.items():\n",
    "                fp = os.path.join(\n",
    "                    EXP2_ROOT,\n",
    "                    f\"fish{fish}_model_{meta['stub']}_Untrained_run{run}_seq{s}_test_preds.npy\"\n",
    "                )\n",
    "                if os.path.isfile(fp):\n",
    "                    preds = np.load(fp)\n",
    "                    rmse_val = np.sqrt(np.mean((preds - y_true) ** 2))\n",
    "                    rmse[s][m][\"untr\"].append(rmse_val)\n",
    "\n",
    "# compute means, errors, statistics\n",
    "means, errs, pvals = {}, {}, {}\n",
    "for s in seq_lengths:\n",
    "    means[s], errs[s], pvals[s] = {}, {}, {}\n",
    "    for m in model_info:\n",
    "        a = np.array(rmse[s][m][\"prtr\"])\n",
    "        b = np.array(rmse[s][m][\"untr\"])\n",
    "        means[s][(m, \"prtr\")] = a.mean() if len(a) > 0 else np.nan\n",
    "        errs[s][(m, \"prtr\")]  = sem(a)       if len(a) > 1 else 0\n",
    "        means[s][(m, \"untr\")] = b.mean() if len(b) > 0 else np.nan\n",
    "        errs[s][(m, \"untr\")]  = sem(b)       if len(b) > 1 else 0\n",
    "        if len(a) == len(b) and len(a) > 0:\n",
    "            pvals[s][m] = wilcoxon(a, b).pvalue\n",
    "        else:\n",
    "            pvals[s][m] = np.nan\n",
    "\n",
    "plt.rcParams[\"pdf.fonttype\"] = 42\n",
    "fig, ax = plt.subplots(figsize=(7, 3), dpi=300)\n",
    "\n",
    "bw = 0.2\n",
    "gap = 0.05\n",
    "offsets = (\n",
    "    np.array([-1.5, -0.5, 0.5, 1.5]) * bw +\n",
    "    np.array([-gap, -gap, gap, gap])\n",
    ")\n",
    "model_names = list(model_info.keys())\n",
    "centers = np.arange(len(model_names)) * ((offsets.max() - offsets.min()) + 0.5)\n",
    "\n",
    "seen = set()\n",
    "for i, m in enumerate(model_names):\n",
    "    center = centers[i]\n",
    "    for j, s in enumerate(seq_lengths):\n",
    "        for k, st in enumerate([\"prtr\", \"untr\"]):\n",
    "            x = center + offsets[j*2 + k]\n",
    "            val = means[s][(m, st)]\n",
    "            err = errs[s][(m, st)]\n",
    "            if np.isnan(val):\n",
    "                continue\n",
    "            col = model_info[m][\"c\"]\n",
    "            if st == \"untr\":\n",
    "                hatch, edgec, lw = \"\", col, 1\n",
    "            else:\n",
    "                hatch, edgec, lw = \"//\", \"white\", 0\n",
    "            label = None\n",
    "            if (m, st) not in seen:\n",
    "                label = f\"{m} {states[st]}\"\n",
    "                seen.add((m, st))\n",
    "            ax.bar(\n",
    "                x, val, yerr=err, width=bw,\n",
    "                color=col, edgecolor=edgec, linewidth=lw,\n",
    "                hatch=hatch, capsize=3, label=label\n",
    "            )\n",
    "\n",
    "ax.set_xticks([])\n",
    "for i, m in enumerate(model_names):\n",
    "    center = centers[i]\n",
    "    ax.text(\n",
    "        center + offsets[:2].mean(), -0.05, r\"$s=5$\",\n",
    "        ha=\"center\", va=\"top\", transform=ax.get_xaxis_transform()\n",
    "    )\n",
    "    ax.text(\n",
    "        center + offsets[2:].mean(), -0.05, r\"$s=20$\",\n",
    "        ha=\"center\", va=\"top\", transform=ax.get_xaxis_transform()\n",
    "    )\n",
    "    ax.text(\n",
    "        center, -0.20, m,\n",
    "        ha=\"center\", va=\"top\", transform=ax.get_xaxis_transform()\n",
    "    )\n",
    "\n",
    "ax.set_ylabel(\"RMSE\")\n",
    "ax.spines[\"top\"].set_visible(False)\n",
    "ax.spines[\"right\"].set_visible(False)\n",
    "ax.set_ylim(0.04, None)\n",
    "\n",
    "style_handles = [\n",
    "    mpatches.Patch(facecolor='white', edgecolor='black',       label='Untrained'),\n",
    "    mpatches.Patch(facecolor='white', edgecolor='black', hatch='//', label='Pretrained'),\n",
    "]\n",
    "ax.legend(\n",
    "    style_handles,\n",
    "    [h.get_label() for h in style_handles],\n",
    "    ncol=2, frameon=False,\n",
    "    bbox_to_anchor=(0.5, -0.3), loc='upper center'\n",
    ")\n",
    "\n",
    "def bracket(x1, x2, y, h):\n",
    "    ax.plot([x1, x1, x2, x2], [y, y+h, y+h, y], c=\"k\", lw=1)\n",
    "    ax.text((x1 + x2)/2, y + h*1.05, \"*\", ha=\"center\", va=\"bottom\")\n",
    "\n",
    "h_br = 0.001\n",
    "for i, m in enumerate(model_names):\n",
    "    center = centers[i]\n",
    "    for j, s in enumerate(seq_lengths):\n",
    "        if pvals[s][m] < 0.05:\n",
    "            top = max(\n",
    "                means[s][(m, \"prtr\")] + errs[s][(m, \"prtr\")],\n",
    "                means[s][(m, \"untr\")] + errs[s][(m, \"untr\")]\n",
    "            )\n",
    "            x1 = center + offsets[j*2]\n",
    "            x2 = center + offsets[j*2 + 1]\n",
    "            bracket(x1, x2, top + h_br, h_br)\n",
    "\n",
    "plt.tight_layout()\n",
    "outfile = os.path.join(EXP2_ROOT, \"pretrained_vs_untrained_exp1_exp2.pdf\")\n",
    "plt.savefig(outfile, dpi=300, transparent=True, bbox_inches=\"tight\")\n",
    "print(\"Saved to\", outfile)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a154d5b6-bda9-49cd-85b5-ee8ed0d42431",
   "metadata": {},
   "source": [
    "# 3) Compare embedding strategies"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "3e597d9f-7e74-4e41-9124-fda03f6f3831",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Pairwise Wilcoxon signed-rank p-values (pooled across models):\n",
      "Vanilla      vs Positional  : p = 4.5216e-01\n",
      "Vanilla      vs RelativePos : p = 3.5988e-01\n",
      "Vanilla      vs Sparse      : p = 6.4083e-01\n",
      "Vanilla      vs Spectral    : p = 5.4253e-01\n",
      "Positional   vs RelativePos : p = 8.7121e-01\n",
      "Positional   vs Sparse      : p = 1.7060e-01\n",
      "Positional   vs Spectral    : p = 8.0783e-01\n",
      "RelativePos  vs Sparse      : p = 3.3874e-01\n",
      "RelativePos  vs Spectral    : p = 6.1201e-01\n",
      "Sparse       vs Spectral    : p = 5.8376e-01\n",
      "Saved plot to: C:\\Users\\jacob\\OneDrive\\Documents\\GitHub\\external_data\\nlp4neuro\\results_april30\\experiment3_results\\final_plots_and_stats\\embeddings_pooled_models.pdf\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 2100x750 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import os\n",
    "import glob\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "from scipy.stats import sem, wilcoxon\n",
    "\n",
    "ROOT = os.path.abspath(\n",
    "    os.path.join(\n",
    "        os.getcwd(), os.pardir, os.pardir, \"results\", \"experiment_3\"\n",
    "    )\n",
    ")\n",
    "\n",
    "RUNS = range(1, 11)\n",
    "EMB_STRATS = [\"Vanilla\", \"Positional\", \"RelativePos\", \"Sparse\", \"Spectral\"]\n",
    "MODEL_INFO = {\n",
    "    \"GPT-2 (base)\":      {\"stub\": \"gpt2\",      \"c\": \"#1f77b4\"},\n",
    "    \"DeepSeek-coder-7b\": {\"stub\": \"deepseek\",  \"c\": \"#000000\"},\n",
    "    \"BERT-small\":        {\"stub\": \"bert\",      \"c\": \"#7f7f7f\"},\n",
    "}\n",
    "\n",
    "# load ground truth\n",
    "for gt in (\"final_predictions_groundtruth_test.npy\", \"groundtruth_test.npy\"):\n",
    "    fp = os.path.join(ROOT, gt)\n",
    "    if os.path.isfile(fp):\n",
    "        y_true = np.load(fp)\n",
    "        break\n",
    "else:\n",
    "    raise FileNotFoundError(\"ground-truth test array not found\")\n",
    "\n",
    "# collect RMSEs\n",
    "rmse = {emb: {m: [] for m in MODEL_INFO} for emb in EMB_STRATS}\n",
    "for run in RUNS:\n",
    "    run_dir = os.path.join(ROOT, f\"run_{run}\")\n",
    "    if not os.path.isdir(run_dir):\n",
    "        continue\n",
    "    for m, info in MODEL_INFO.items():\n",
    "        fam_dir = os.path.join(run_dir, f\"{info['stub']}_embedding_comparisons\")\n",
    "        for emb in EMB_STRATS:\n",
    "            pred_fp = os.path.join(\n",
    "                fam_dir,\n",
    "                emb.lower(),\n",
    "                f\"{info['stub']}_{emb.lower()}_preds_run{run}.npy\"\n",
    "            )\n",
    "            if os.path.isfile(pred_fp):\n",
    "                preds = np.load(pred_fp)\n",
    "                rmse_val = np.sqrt(np.mean((preds - y_true) ** 2))\n",
    "                rmse[emb][m].append(rmse_val)\n",
    "\n",
    "# pooled statistics\n",
    "pooled = {\n",
    "    emb: np.concatenate([rmse[emb][m] for m in MODEL_INFO])\n",
    "    for emb in EMB_STRATS\n",
    "}\n",
    "pooled_means = {emb: pooled[emb].mean() for emb in EMB_STRATS}\n",
    "pooled_sems  = {emb: sem(pooled[emb])      for emb in EMB_STRATS}\n",
    "\n",
    "model_means = {\n",
    "    emb: {m: np.mean(rmse[emb][m]) for m in MODEL_INFO}\n",
    "    for emb in EMB_STRATS\n",
    "}\n",
    "\n",
    "# plotting\n",
    "plt.rcParams[\"pdf.fonttype\"] = 42\n",
    "fig, ax = plt.subplots(figsize=(7, 2.5), dpi=300)\n",
    "\n",
    "x  = np.arange(len(EMB_STRATS))\n",
    "bw = 0.6\n",
    "\n",
    "ax.bar(\n",
    "    x,\n",
    "    [pooled_means[e] for e in EMB_STRATS],\n",
    "    yerr=[pooled_sems[e] for e in EMB_STRATS],\n",
    "    capsize=4,\n",
    "    width=bw,\n",
    "    color=\"lightgray\",\n",
    "    edgecolor=\"none\"\n",
    ")\n",
    "\n",
    "n_models = len(MODEL_INFO)\n",
    "offsets = np.linspace(-bw/3, bw/3, n_models)\n",
    "for j, (m, info) in enumerate(MODEL_INFO.items()):\n",
    "    xs = x + offsets[j]\n",
    "    ys = [model_means[emb][m] for emb in EMB_STRATS]\n",
    "    ax.scatter(xs, ys, color=info[\"c\"], label=m, zorder=5, s=40)\n",
    "\n",
    "ax.set_xticks(x)\n",
    "ax.set_xticklabels(EMB_STRATS, rotation=15)\n",
    "ax.set_ylim(0.04, None)\n",
    "ax.set_ylabel(\"RMSE\")\n",
    "ax.spines[\"top\"].set_visible(False)\n",
    "ax.spines[\"right\"].set_visible(False)\n",
    "\n",
    "fig.subplots_adjust(bottom=0.09)\n",
    "ax.legend(\n",
    "    ncol=len(MODEL_INFO),\n",
    "    frameon=False,\n",
    "    bbox_to_anchor=(0.5, -0.3),\n",
    "    loc=\"upper center\"\n",
    ")\n",
    "\n",
    "print(\"\\nPairwise Wilcoxon signed-rank p-values (pooled across models):\")\n",
    "for i in range(len(EMB_STRATS)):\n",
    "    for j in range(i+1, len(EMB_STRATS)):\n",
    "        e1, e2 = EMB_STRATS[i], EMB_STRATS[j]\n",
    "        data1, data2 = pooled[e1], pooled[e2]\n",
    "        if len(data1) == len(data2) and len(data1) > 0:\n",
    "            stat, p = wilcoxon(data1, data2)\n",
    "            print(f\"{e1:12s} vs {e2:12s}: p = {p:.4e}\")\n",
    "        else:\n",
    "            print(f\"{e1:12s} vs {e2:12s}: insufficient data\")\n",
    "\n",
    "out = os.path.join(ROOT, \"embeddings_pooled_models.pdf\")\n",
    "plt.tight_layout()\n",
    "plt.savefig(out, dpi=300, transparent=True, bbox_inches=\"tight\")\n",
    "print(\"Saved plot to:\", out)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "57309e2b-2f0b-4b1c-9b43-45eca8c292b2",
   "metadata": {},
   "source": [
    "## 3c)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 136,
   "id": "02becd0f-ecf7-4c13-8259-f0003730b489",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Figure saved to: figures\\barplot_random_by_model.pdf\n",
      "\n",
      "Pooled-transformer Mann–Whitney U p-values  (condition = random):\n",
      "Vanilla     vs Positional : p = 8.883e-01\n",
      "Vanilla     vs RelativePos: p = 1.260e-01\n",
      "Vanilla     vs Sparse     : p = 2.939e-01\n",
      "Vanilla     vs Spectral   : p = 5.201e-01\n",
      "Positional  vs RelativePos: p = 2.398e-01\n",
      "Positional  vs Sparse     : p = 2.116e-01\n",
      "Positional  vs Spectral   : p = 5.793e-01\n",
      "RelativePos vs Sparse     : p = 1.628e-02\n",
      "RelativePos vs Spectral   : p = 5.493e-01\n",
      "Sparse      vs Spectral   : p = 1.055e-01\n",
      "\n",
      "Within-model Mann–Whitney U p-values:\n",
      "\n",
      "  GPT2:\n",
      "    Vanilla     vs Positional : p = 9.698e-01 \n",
      "    Vanilla     vs RelativePos: p = 1.726e-02 **\n",
      "    Vanilla     vs Sparse     : p = 2.263e-01 \n",
      "    Vanilla     vs Spectral   : p = 7.337e-01 \n",
      "    Positional  vs RelativePos: p = 1.133e-02 **\n",
      "    Positional  vs Sparse     : p = 7.566e-02 \n",
      "    Positional  vs Spectral   : p = 9.698e-01 \n",
      "    RelativePos vs Sparse     : p = 3.298e-04 **\n",
      "    RelativePos vs Spectral   : p = 1.402e-02 **\n",
      "    Sparse      vs Spectral   : p = 1.859e-01 \n",
      "\n",
      "  DEEPSEEK:\n",
      "    Vanilla     vs Positional : p = 9.097e-01 \n",
      "    Vanilla     vs RelativePos: p = 1.000e+00 \n",
      "    Vanilla     vs Sparse     : p = 5.708e-01 \n",
      "    Vanilla     vs Spectral   : p = 5.708e-01 \n",
      "    Positional  vs RelativePos: p = 6.776e-01 \n",
      "    Positional  vs Sparse     : p = 5.205e-01 \n",
      "    Positional  vs Spectral   : p = 4.727e-01 \n",
      "    RelativePos vs Sparse     : p = 9.698e-01 \n",
      "    RelativePos vs Spectral   : p = 4.274e-01 \n",
      "    Sparse      vs Spectral   : p = 2.123e-01 \n",
      "\n",
      "  BERT:\n",
      "    Vanilla     vs Positional : p = 4.727e-01 \n",
      "    Vanilla     vs RelativePos: p = 3.847e-01 \n",
      "    Vanilla     vs Sparse     : p = 3.447e-01 \n",
      "    Vanilla     vs Spectral   : p = 3.075e-01 \n",
      "    Positional  vs RelativePos: p = 6.776e-01 \n",
      "    Positional  vs Sparse     : p = 1.212e-01 \n",
      "    Positional  vs Spectral   : p = 6.232e-01 \n",
      "    RelativePos vs Sparse     : p = 4.515e-02 ****\n",
      "    RelativePos vs Spectral   : p = 2.123e-01 \n",
      "    Sparse      vs Spectral   : p = 5.390e-02 \n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 2700x750 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import re\n",
    "import itertools\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "from scipy.stats import mannwhitneyu, sem\n",
    "from pathlib import Path\n",
    "\n",
    "LOGFILE = \"salient_random_logs.txt\"\n",
    "\n",
    "OUTDIR = os.path.abspath(\n",
    "    os.path.join(\n",
    "        os.getcwd(), os.pardir, os.pardir, \"results\", \"experiment_3\"\n",
    "    )\n",
    ")\n",
    "\n",
    "EMBEDS = [\"Vanilla\", \"Positional\", \"RelativePos\", \"Sparse\", \"Spectral\"]\n",
    "MODELS = [\"gpt2\", \"deepseek\", \"bert\"]\n",
    "COLORS = {\n",
    "    \"gpt2\": \"#1f77b4\",\n",
    "    \"deepseek\": \"#000000\",\n",
    "    \"bert\": \"#7f7f7f\"\n",
    "}\n",
    "DISPLAY = {\n",
    "    \"gpt2\": \"GPT-2\",\n",
    "    \"deepseek\": \"DeepSeek-c7b\",\n",
    "    \"bert\": \"BERT-bu\"\n",
    "}\n",
    "COND_TO_PLOT = \"random\"\n",
    "alpha = 0.05\n",
    "\n",
    "# now match just the plain text marker\n",
    "cond_pat = re.compile(r\"(salient50_removed|random50_removed)\")\n",
    "\n",
    "# now match lines like \"DeepSeek + Vanilla\" (possibly with leading spaces)\n",
    "arrow_pat = re.compile(r\"^\\s*([A-Za-z0-9]+)\\s+\\+\\s+([A-Za-z0-9]+)\")\n",
    "\n",
    "# unchanged, still matches lines like \"   RMSE 0.04133\"\n",
    "summary_pat = re.compile(r\"^\\s+RMSE\\s+([0-9.]+)$\")\n",
    "\n",
    "by_mod = {\n",
    "    cond: {\n",
    "        emb: {model: [] for model in MODELS}\n",
    "        for emb in EMBEDS\n",
    "    }\n",
    "    for cond in (\"salient\", \"random\")\n",
    "}\n",
    "\n",
    "cur_cond = cur_model = cur_emb = None\n",
    "with open(LOGFILE) as f:\n",
    "    for line in f:\n",
    "        m = cond_pat.search(line)\n",
    "        if m:\n",
    "            cur_cond = \"salient\" if \"salient\" in m.group(1) else \"random\"\n",
    "            continue\n",
    "        m = arrow_pat.search(line)\n",
    "        if m:\n",
    "            cur_model = m.group(1).lower()\n",
    "            raw_emb = m.group(2)\n",
    "            cur_emb = \"RelativePos\" if raw_emb.lower() == \"relativepos\" else raw_emb.capitalize()\n",
    "            continue\n",
    "        m = summary_pat.match(line)\n",
    "        if m and cur_cond and cur_model and cur_emb:\n",
    "            by_mod[cur_cond][cur_emb][cur_model].append(float(m.group(1)))\n",
    "\n",
    "means = np.array([\n",
    "    [np.mean(by_mod[COND_TO_PLOT][emb][model]) for model in MODELS]\n",
    "    for emb in EMBEDS\n",
    "])\n",
    "errs = np.array([\n",
    "    [sem(by_mod[COND_TO_PLOT][emb][model]) for model in MODELS]\n",
    "    for emb in EMBEDS\n",
    "])\n",
    "pooled = {\n",
    "    emb: sum((by_mod[COND_TO_PLOT][emb][m] for m in MODELS), [])\n",
    "    for emb in EMBEDS\n",
    "}\n",
    "\n",
    "pair_pvals = {}\n",
    "for i, j in itertools.combinations(range(len(EMBEDS)), 2):\n",
    "    _, p = mannwhitneyu(\n",
    "        pooled[EMBEDS[i]],\n",
    "        pooled[EMBEDS[j]],\n",
    "        alternative=\"two-sided\"\n",
    "    )\n",
    "    pair_pvals[(i, j)] = p\n",
    "\n",
    "fig, ax = plt.subplots(figsize=(9, 2.5), dpi=300)\n",
    "x = np.arange(len(EMBEDS))\n",
    "bw = 0.22\n",
    "\n",
    "for m_idx, model in enumerate(MODELS):\n",
    "    xs = x + (m_idx - 1) * bw\n",
    "    ax.bar(\n",
    "        xs,\n",
    "        means[:, m_idx],\n",
    "        yerr=errs[:, m_idx],\n",
    "        width=bw,\n",
    "        color=COLORS[model],\n",
    "        capsize=3,\n",
    "        label=DISPLAY[model]\n",
    "    )\n",
    "\n",
    "ax.set_xticks(x)\n",
    "ax.set_xticklabels(\n",
    "    [\"Linear\", \"Pos.\", \"Rel. Pos.\", \"Sparse AE\", \"Spectral\"],\n",
    "    rotation=15\n",
    ")\n",
    "ax.set_ylabel(\"RMSE\")\n",
    "ax.spines[\"top\"].set_visible(False)\n",
    "ax.spines[\"right\"].set_visible(False)\n",
    "ax.legend(\n",
    "    ncol=3,\n",
    "    frameon=False,\n",
    "    loc=\"upper center\",\n",
    "    bbox_to_anchor=(0.5, -0.20)\n",
    ")\n",
    "\n",
    "h = 0.002\n",
    "pad = -h * 6\n",
    "stack_sep = h * 2.5\n",
    "\n",
    "tops_emb = [max(means[i] + errs[i]) for i in range(len(EMBEDS))]\n",
    "pooled_stack = 0\n",
    "\n",
    "for (i, j), p in pair_pvals.items():\n",
    "    if p < alpha:\n",
    "        y0 = max(tops_emb[i], tops_emb[j])\n",
    "        y = y0 + h + pooled_stack * stack_sep\n",
    "\n",
    "        x0 = i - 1.2 * bw\n",
    "        x1 = j + 1.2 * bw\n",
    "\n",
    "        ax.plot([x0, x0, x1, x1],\n",
    "                [y, y + h, y + h, y],\n",
    "                color=\"k\", lw=1)\n",
    "        ax.text(\n",
    "            (x0 + x1) / 2,\n",
    "            y + h + pad,\n",
    "            \"*\",\n",
    "            ha=\"center\", va=\"bottom\", color=\"k\"\n",
    "        )\n",
    "\n",
    "        pooled_stack += 1\n",
    "\n",
    "base_offset = pooled_stack + 5\n",
    "stacks_by_model = {m_idx: 0 for m_idx in range(len(MODELS))}\n",
    "star_map = {0: \"**\", 1: \"***\", 2: \"****\"}\n",
    "\n",
    "for m_idx, model in enumerate(MODELS):\n",
    "    for i, j in itertools.combinations(range(len(EMBEDS)), 2):\n",
    "        data_i = by_mod[COND_TO_PLOT][EMBEDS[i]][model]\n",
    "        data_j = by_mod[COND_TO_PLOT][EMBEDS[j]][model]\n",
    "        _, p = mannwhitneyu(data_i, data_j, alternative=\"two-sided\")\n",
    "        if p < alpha:\n",
    "            x0 = i + (m_idx - 1) * bw\n",
    "            x1 = j + (m_idx - 1) * bw\n",
    "\n",
    "            top_i = means[i, m_idx] + errs[i, m_idx]\n",
    "            top_j = means[j, m_idx] + errs[j, m_idx]\n",
    "            y0 = max(top_i, top_j)\n",
    "\n",
    "            model_base = base_offset + (m_idx * 8.5)\n",
    "            offset = model_base + stacks_by_model[m_idx]\n",
    "            y = y0 + h + offset * stack_sep\n",
    "\n",
    "            ax.plot([x0, x0, x1, x1],\n",
    "                    [y, y + h, y + h, y],\n",
    "                    color=\"k\", lw=1)\n",
    "            ax.text(\n",
    "                (x0 + x1) / 2,\n",
    "                y + h + pad,\n",
    "                star_map[m_idx],\n",
    "                ha=\"center\", va=\"bottom\", color=\"k\"\n",
    "            )\n",
    "\n",
    "            stacks_by_model[m_idx] += 1\n",
    "\n",
    "fig.tight_layout()\n",
    "out_pdf = OUTDIR / \"barplot_random_by_model.pdf\"\n",
    "fig.savefig(out_pdf, transparent=True, bbox_inches=\"tight\")\n",
    "print(\"Figure saved to:\", out_pdf)\n",
    "\n",
    "print(f\"\\nPooled-transformer Mann–Whitney U p-values  (condition = {COND_TO_PLOT}):\")\n",
    "for (i, j), p in pair_pvals.items():\n",
    "    print(f\"{EMBEDS[i]:11s} vs {EMBEDS[j]:11s}: p = {p:.3e}\")\n",
    "\n",
    "print(\"\\nWithin-model Mann–Whitney U p-values:\")\n",
    "for m_idx, model in enumerate(MODELS):\n",
    "    print(f\"\\n  {model.upper()}:\")\n",
    "    for i, j in itertools.combinations(range(len(EMBEDS)), 2):\n",
    "        _, p = mannwhitneyu(\n",
    "            by_mod[COND_TO_PLOT][EMBEDS[i]][model],\n",
    "            by_mod[COND_TO_PLOT][EMBEDS[j]][model],\n",
    "            alternative=\"two-sided\"\n",
    "        )\n",
    "        sig = star_map[m_idx] if p < alpha else \"\"\n",
    "        print(f\"    {EMBEDS[i]:11s} vs {EMBEDS[j]:11s}: p = {p:.3e} {sig}\")\n"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.13.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}
